SUfTlARY AHD AflALYSIS OF COfTIBITS
ON THE
NOTICE OF PROTOSED RULB1AKING
FOR
GASEOUS B1ISSION REGULATIONS
FOR 1983 AND LATER MODEL YEAR LIGHT-DUTY TRUCKS
BMIWEfiAL PROTECTION AGENCY
OFFICE OF AIR, NOISE, AND RADIATION
OFFICE OF MOBILE SOURCE AIR POLLUTION COffl^OL
f-1AY 1990
-------
EPA-420-R-80-101
SUMMARY AND ANALYSIS OF COMMENTS
ON THE
NOTICE OF PROPOSED RULEMAKING
FOR
GASEOUS EMISSION REGULATIONS
FOR 1983 AND LATER MODEL YEAR LIGHT-DUTY TRUCKS
MAY 1980
STANDARDS DEVELOPMENT AND SUPPORT BRANCH
EMISSION CONTROL TECHNOLOGY DIVISION
OFFICE OF MOBILE SOURCE AIR POLLUTION CONTROL
OFFICE OF AIR, NOISE, AND RADIATION
U.S. ENVIRONMENTAL PROTECTION AGENCY
-------
Important Notice
At the time we performed the analyses found in this document
the rulemaking was intended to be applicable to the 1983 model
year. In the very late stages of the rulemaking, EPA decided to
delay the requirements of the regulations until 1984. Since the
conclusions we reached remain virtually the same for 1984, the
analyses and recommendations have not been changed. For this
reason, the model year 1983 is used as a reference point thorughout
the document.
-------
Table of Contents
Page
I. Introduction ii
II. List of Commenters iii
III. Analysis of Issues /
A. Redefinition of "Useful Life" |
B. In-Use Durability Testing ^
C. Allowable Maintenance 7
D. Idle Test and Standards /V
t
E. Leadtime
F. Economic Impact
G. Technological Feasibility 7 £
H. Selective Enforcement Auditing
I. Nonconformance Penalty
J. Diesel Crankcase Emissions Control
K. Numerical Standards/Standards Derivation ....
L. Fuel Economy
M. Environmental Impact
N. Special Exemptions
-------
I. Introduction
The Environmental Protection Agency (EPA) published a Notice
of Proposed Rulemaking (NPRM) on Thursday, July 12, 1979, proposing
new light-duty truck emissions regulations for 1983 and later model
years. The proposed rule prescribed more stringent hydrocarbon and
carbon monoxide emission standards, and established a revised
assembly-line testing program and nonconformance penalty system for
1983 and later model year light-duty trucks as mandated by the
Clean Air Act Amendments of 1977. Substantial changes were also
proposed to the definition of useful life, and the procedures used
to verify the durability of emission control systems over their
useful life. Crankcase emission standards for diesel engines were
also proposed.
This document presents a summary and analysis of comments
received in response to the NPRM.
11
-------
II.
1.
2.
3.
4.
5.
6.
7.
8.
S.
10.
11.
12.
13.
14.
15.
16.
17.
18.
List of Commenters
Alaska Department of Environmental Conservation
American Motors
Glen F. Brammeier, NHTSA, DOT
Robert Chivvis, EPA Alaska Air Coordinator
Chrysler Corporation
Cummins Engine Company
Council on Wage and Price Stability
United States Department of Commerce
Ford Motor Company
General Motors Corporation
Doug Hansen, Director, Air and Hazardous Materials Div. ,
EPA
International Harvester Company
Motor Vehicle Manufacturers Association
Municipality of Anchorage, Alaska
New Mexico Cattle Growers Association
Toyo Kogyo Co, Ltd. (Mazda)
Toyota Motor Company, Ltd.
Volkswagen of America, Inc.
ADEC
AM
Chrysler
Cummins
COWPS
DOC
Ford
GM
IH/IHC
MVMA
Toyo Kogyo
Toyota
Volkswagen
111
-------
III. Analysis of Issues
-------
A. Issue: Redefinition of Useful Life
1. Summary of the Issue
In the July 12, 1979 NPRM, EPA proposed to amend the current
definition of "useful life" for light-duty trucks. Currently,
light-duty truck useful life is defined as five years or 50,000
miles (or the equivalent) whichever occurs first. This definition,
which applies to all light-duty trucks, was found to be inadequate
due to deterioration of the emission controls of vehicles that are
beyond 5 years or 50,000 miles. The proposed revision defines the
useful life as the "average period of use up to the vehicle retire-
ment or engine replacement or rebuild." Since this period may vary
among manufacturers, and among vehicle types produced by a single
manufacturer, EPA proposes that the manufacturers of a vehicle
determine the duration of this period. The useful life is, however,
constrained to be not less than five years or 50,000 miles which-
ever comes first, or the period of the.,basic mechanical warranty on
the engine assembly, whichever is longer.
2. Summary of the Comments
A large number of comments were received regarding the EPA's
justification for changing the useful life definition by noting
that the half-life concept has been a part of vehicle emission
regulations since the 1966 (HEW) rules applying to 1968 model year
vehicles. In that rulemaking, 100,000 miles was defined as the
basis for "lifetime emissions." Under the assumption that emission
deteriorations would be linear, HEW established a procedure for
calculating average lifetime emissions at the approximate half-life
(50,000-mile) point. All subsequent regulations for light-duty
vehicles, light-duty trucks, and heavy-duty engines, except for the
recently adopted heavy-duty engine regulations, have used half
of the expected life as the useful life. The commenters imply that
the average lifetime emissions concept has embodied the intentions
of Congress throughout the years and that the proposed redefinition
of the useful life directly contravenes Congressional intent. They
contend that Congress in drawing up the 1970 Clean Air Act (where
the term "useful life" first appears) was fully aware that the
50,000 mile "lifetime" they chose for durability and warranty
purposes approximated only half of the expected life of a light-
duty vehicle. The half-life concept was therefore specifically
woven into the Act. In further support of these contentions one
commenter states that an analysis of legislative history of the
1977 Clean Air Act Amendments clearly shows a rejection of the
total life concept now being proposed by EPA.
Most commenters argued against the concept of a full-life
useful life on the basis that it acts to increase the stringency of
the emission standards. As the light-duty truck regulations are
presently constructed, manufacturers must design their engines so
that during approximately the first half of the lifetime their
emissions do not deteriorate past tke_level of the standards.
-------
i
It was argued that this situation requires the emissions of a new
engine to be somewhat below the standard in order that deteriora-
tion may be accommodated. The proposed full life concept would
purportedly require lifelong emissions compliance and hence shall
require lower low mileage targets (LMT's). This is the "increased
stringency" referred to in the comments. Some of the commenters
went on to claim that EPA is in effect requiring emission reduc-
tions in excess of the 90 percent set by Congress.
The two remaining major areas of comment were directed at
specific problems which were expected to arise during implemen-
tation of the full-life concept. The first of these is the lan-
\ guage of the NPRM that requires the manufacturers to determine an
•^"average" period of engine use for each engine line. The comments
imply that half of the engines subject to an average useful life
will require rebuild or retirement before they reach that useful
life. Most commenters said a flurry of warranty claims could be
expected to result from decay in emission-related components toward
the end of the useful life.
The last set of comments were procedural in nature and cen-
tered around the difficulties that the manufacturers would expect
in defining a useful life number under the proposed full-life
concept. First, data concerning actual engine usage periods is
largely unavailable at this time. Secondly, it was argued that
the decision when to retire or rebuild is reached by the user on
largely economic, as opposed to mechanical grounds. Thus, manu-
facturers would find it difficult to arrive at an average period
for this event for an engine. The problem would be further
compounded by the wide range of vocational applications seen by
many engine families.
3. Analysis of the Comments
Many of the issues concerning useful life were previously
raised in connection with the 1984 heavy-duty engine gaseous
emissions rulemaking. These issues are analyzed in depth in the
Summary and Analysis of Comments document accompanying the final
heavy-duty rulemaking, and are incorporated herin by reference.
\ The Clean Air Act did not place a half-life constraint on
light-duty trucks' useful lives. Quite to the contrary, Section
202(d)(2) clearly provides the Administrator with the discretionary
authority to define the useful life of light-duty trucks as
greater than that set by Congress for light-duty vehicles if he
"determines that a period of use of greater duration or mileage
is appropriate." Given the disireability of manufacturers'
building emission control components as durable as the rest of the
traditionally long-lasting engine parts, and given the significant
air quality benefits that will be realized if the proposed defin-
ition of useful life is adopted, adoption of the full useful life
concept is appropriate and well within the discretionary authority
explicitly granted to the Administrator by the Act.
-------
Legislative history shows no evidence of a Congressional '
commitment to impose a half-life limit on the useful life of
heavy-duty vehicles or light-duty trucks. While the 1977 Clean Air
Act Amendments deliberately defined light-duty vehicle useful life
as half of the expected actual life, that decision was a result of
forces that were present at that specific time with respect to that
specific class of vehicles. A 100,000-mile/10-year requirement was
seriously considered by a Senate Committee, but was subsequently
halved based upon the EPA staff's analysis of economic and tech-
nological feasibility. Compliance with full useful life standards
is indeed possible and economically feasible in 1983 for light-duty
trucks. (See Section 1 of this Summary and Analysis of Comments).
Several comments implied that past regulatory practice
should constrain future rulemaking. It is EPA's belief, however,
that regulations must be the best attempt possible at any given
time to fulfill the wishes of Congress within the context of
feasibility, cost, and other factors. t It was such constraints as
these which initially resulted in use of a half-life useful life by
HEW; and which led to EPA's recent adoption of full life useful
life for heavy-duty trucks in the rulemaking for 1984 heavy-duty
engine gaseous emissions.
The final area of comment which affects the full-life concept
itself is the stringency issue. This idea might be better treated
in the broad context of how it fits into the total full-life
useful life plan. Given the fact that the Administrator has the
authority to adopt a full-life useful life, then a lower zero-mile*/
emission level is simply a practical result of applying that
requirement to the certification process. Thus, we agree that in a
narrow sense, the design-goal emission level is more stringent
under a full useful life concept. However, the standards them-
selves are not more stringent; they are simply met for the lifetime
of the engine. The staff cannot accept the stringency issue as an
argument against the full-life useful life. In any event, Congress
required standards representing a reduction of "at least 90 per- >'
cent", thus indicating that more stringent standards could be
promulgated, provided they are technologically feasible.
The remaining discussion will deal with the practical diffi-
culties associated with the full-life useful life concept. The
first of those is the proposed requirement that the useful life
value supplied by the manufacturer be the "average" for that engine
family. The staff has considered alternative methods of establish-
ing the useful life number, though no commenter offered suggestions
along those lines. For example, the alternative of allowing
complete latitude in defining the useful life is likely to encour-
age unrepresentative values. There would be a somewhat natural
trend for manufacturers, favoring short durability programs and few
warranty claims, to gravitate toward a lower useful life limit.
Another alternative could be for EPA to establish that some percen-
tile of the retirement/rebuild distribution be used instead of a
straight average. This option, however^ suffers from a complete
lack of data to support the selection of any specific percentile.
-------
J
The staff has concluded that the useful life value supplied by
the manufacturers should be an "average" for a given engine
family. This appears to be the best way to determine useful life
keeping both industry and EPA's best interest in mind.
Regarding the large number of warranty claims anticipated by
the commenters, the staff disagrees that the proposed rule will
result in half of the manufacturers' engines requiring emissions
warranty work. Although it is clear that half will reach their
individual retirement/rebuild points, this does not necessarily
mean an emissions violation will exist in every case. Certainly
there could be additional warranty claims attributable to the
extension of the useful life period. The staff does not, however,
expect this number to be substantial. Rather, we expect manufac-
turers to make needed changes in components to provide full life
durability equivalent to their present 50,000 mile durability.
Costs for these changes have been included in the economic analysis
of the rulemaking. Additionally, it was commented that the pro-
posed regulations imply that the manufacturer would be responsible
for post-rebuild emissions compliance. To alleviate this problem
the staff recommends a change in the proposed rule which defines
the end of an engine's useful life as the average period of use or
the point at which the engine needs rebuilding, whichever is
reached first (provided that the 50,000 mile/5 year minimum has
been passed). Thus, the cost of rebuild, as well as all subsequent
repairs, will be borne by the owner and not the manufacturer.
4. Staff Recommendations
On the basis of comments and their analysis above, the staff
recommends that the useful-life provisions as proposed be retained
largely intact. Three significant changes are offered, however,
which respond to a wide range of comments.
As we concluded during the discussion above, the staff be-
lieves that the full-life useful life concept should remain a part
of this Rulemaking. Within the context, we advocate that the
language "average period of use" be kept intact for the sake of
practicality. Since the manufacturers will be setting the useful
life values, EPA's requiring that value to be an average appears to
be the most reasonable method of encouraging accurate useful
lives.
Several of the difficulties associated with an "average"
useful life, however, will be reduced or eliminated if certain
staff recommendations are adopted. Specifically, we support 1) a
set of more objective criteria for determining when rebuild is
necessary, 2) a manufacturers' option to supply for the owner
alternate expected useful lives depending on service application,
and 3) modifying the "useful life" definition to be less restric-
tive of the manner in which the useful life is determined.
The first of these suggested changes is the most significant
and would remove much of the uncertainty in defining an "average
-------
period of use up to engine retirement or rebuild." The major
criterion for determining whether an engine is due for a rebuild /
would appear on the label and would be, for the purposes of this
rulemaking, a compression test, along with a measure of oil con-
sumption and of bearing failure. Those tests will cover nearly all
mechanical situations which normally signal the need for a rebuild.
Since the actual test values will be determined by the manufacturer
for each engine family, establishing the average useful life should
be easier and more acurate. Another implication is that an "actual
useful life" will exist for each individual engine; there will
be a measureable endpoint to the manufacturer's obligation for
an engine with respect to both durability testing and the emissions
warranty. Thus, the regulations clearly will not require post-
rebuild emissions compliance.
The second recommendation amounts to allowing a qualifying ^
statement on the label to indicate to the owner that the useful
life of this particular engine can be expected to vary from the
"average" due to a lighter or heavier service application. The
label could also direct the reader to the operator's manual for
information about vocation-specific average useful lives, about how """
the emissions-related warranty differs from the mechanical war-
ranty, etc. The purpose of the label change is to promote user
understanding of the "average useful life" concept and hence to
reduce the threat of warranty conflicts.
The final recommendation is to remove from the definition.of
useful life the restriction that for new engines the useful life be
determined from durability testing. We see this provision as an
unnecessary complication of the process of establishing a useful
life value.
Some of our recommedations, particularly the first two, will
to a certain extent add to the complexity of portions of the
regulations and the certification process as compared to the
original proposal. However, the staff is firmly convinced that by
making these adjustments to the proposal, EPA will not only answer
a range of reasonable comments but will improve the workability,
versatility, and fairness of the full-life useful life concept. We
urge the adoption of these provisions.
-------
B. Issue: In-Use Durability Testing
In order to better respond to comments on the proposed in-use
durability testing procedure and to optimize all components of the
program, EPA is delaying the finalization of the in-use durability
testing requirements. Further analysis of the design of the
durability program will continue and finalization of the program is
expected to occur at the same time as the statutory NOx reduction.
The Summary and Analysis of Comments on this component of the
proposal are not included in this document. Instead, comments
received will be addressed when the in-use durability regulations
are finalized.
Beginning in 1984, and continuing until finalization of a
revised durability testing procedure, the burden of durability
testing will be shifted to the manufacturers. Under this concept,
the manufacturers will determine their deterioration factors in
programs which they design. EPA will not approve the programs
which the manufacturers design but will require that they: 1)
describe their durability testing program in the certification
application, 2) certify that their durability testing procedures
account for deterioration of emission related components and other
critical deterioration processes, and 3) adhere to the maintenance
requirements as applicable specified in the allowable maintenance
regulations. These requirements are the same as those proposed for
the determination of the preliminary deterioration factor.
Manufacturers are encouraged to begin small-scale in-use
durability programs in the near future so they can gain some
meaningful experience with in-use durability testing. This will
benefit the manufacturers and EPA in that they could generate
in-use durability data which could verify the feasibility of and
need for an in-use type durability testing program.
-------
C. Issue: Allowable Maintenance
1. Summary of the Issue
Included in Che NPRM were newly-proposed provisions to
limit the amount of maintenance which can be performed on light-
duty truck durability-data engines. Emission-related maintenance
must be technologically necessary and must have a reasonable
likelihood of being performed by owners in the field. Specific
minimum maintenance intervals are proposed which EPA has determined
to be technologically feasible. Additionally, "emission-related
maintenance" and "non-emission-related maintenance" are defined.
These provisions will help ensure that in-use engines do not exceed
the emission standards as a result of control technology which
requires more frequent maintenance than the users will actually
perform.
,t
2. Summary of the Comments
Many of the comments received are similar to the comments
given on the allowable maintenance provisions of the Proposed 1983
Heavy—Duty Engine Regulations.^/ The three categories which
encompass most of the comments are: 1) concerns over EPA's justi-
fication, both legal and logical, for imposing maintenance restric-
tions, 2) criticism of certain of the maintenance intervals, and 3)
comments on the four criteria for assuring "a resonable likelihood
of maintenance being performed in-use."
Beginning with the legal issues, several commenters questioned
EPA's authority to establish "technologically necessary1' intervals
for maintenance. The commenters1 interpretations of §207(c)(3)(A)
and §206(d) of the amended Clean Air Act (CAA) cited in the NPBM
as the basis of the provisions differed from the interpretation of
the agency. IHC interprets §207(c)(3) (A) to mean that the manu-
facturer shall furnish written instructions for the proper main-
tenance of the vehicle and does not indicate that the Admini-
strator is to make any decisions as to what constitutes proper
maintenance. Ford commented that EPA's scheme for controlling
engine maintenance would introduce an "impermissible degree of
uncertainty into the emission certification process." Ford felt
that the manufacturers would have no means to ascertain in advance
of their certification applications whether EPA would classify
engine maintenance operations as "emission related" or not. This
degree of uncertainty is unlawful, Ford states, according to the
recent Paccar decision,^/ in which the court noted:
"Manufacturers are entitled to testing criteria that they can
rely upon with certainty."
In addition to the legal criticisms commenters questioned the
logical and factual basis of EPA's proposed revisions. General
Motors criticized the substitution ofvEPA's judgement as to what
maintenance intervals are necessary in lieu of the manufacturer
-------
recommending maintenance intervals to the customers. The basis for
criticism is that the reduction of vehicle maintenance has been and
will continue to be a goal of the manufacturer. Discussing the
manufacturer's role in recommending maintenance intervals AMC
stated, "Restricting maintenance is counter-productive to air
quality goals and the manufacturer' should have the final say on
what recommended maintenance schedules should be followed by the
owners." AMC went on to say that requiring the manufacturer to
demonstrate the reasonable likelihood of in-use performance of
maintenance is totally unrealistic and the fact that maintenance
may not be performed does not mean that it is unnecessary. Both
Ford and IHC stated that the maintenance schedule, as set up by the
manufacturer, is part of the competitive process in marketing
because the customer is constantly demanding a product which
requires less maintenance. Allowing EPA to recommend maintenance
intervals would destroy an incentive to minimize required mainte-
nance.
One manufacturer commented specifically on the definitions
contained within the proposal as being illogical. Ford criticized
the definitions of "emission related" and "non-emission related"
maintenance and "new technology" as being inadequate to inform
manufacturers whether specific maintenance operations, in addition
to those listed in the NPRM, may be performed at intervals rec-
ommended by the manufacturer, or not at all. The "new technology"
definition is objected to as being a roadblock for a manufacturer
to introduce technology which is new to the manufacturer. In
applying this technology, if it were deemed by EPA as emission
related, the manufacturer would be precluded from performing reason-
able maintenance.
Basically all manufacturers that commented responded that the
technology does not exist to meet the proposed interval require-
ments. Toyota commented that they are not confident that they can
comply with the 1980 model year California light-duty vehicle
maintenance intervals which are less stringent than the EPA pro-
posed intervals. AMC is not aware of any proven technologies which
reflect with any degree of certainty that oxygen sensors will last
for 50,000 miles or that catalysts will last for 130,000. Ford
further criticized EPA for failure to outline either a description
of the improvements required or the basis for its conclusion that
such improvements are feasible.
Commenters expressed concern that EPA has arbitrarily and
incorrectly fixed the maintenance intervals because there is not
any data upon which EPA may reliably predict that all 1983 and
later light-duty trucks will require emission-related maintenance
no more frequently than the specified intervals. Cummins criti-
cized EPA for setting maintenance intervals exactly the same as the
heavy-duty diesel engines. Toyota indicated that EPA did not
describe in the proposal any basis for lengthening the intervals.
Toyota claimed it does not have enough data or experience in the
field to meet the proposed intervals and can only assure perfor-
mance with the current maintenance services.
-------
A substantial volume of comment material was directed at the
more technical issue of the proposed intervals themselves.
For gasoline engines, comments concentrated on the intervals
proposed for spark plug and catalyst replacement; on diesels, the
comments addressed the turbocharger and injector maintenance
intervals. Other areas of comment included the oxygen sensor and
EGR system.
The proposed 30,000-mile maintenance interval for spark plugs
was criticized as being unjustified and unrealistic. GM had
concern about spark plug life in higher mileage engines where a
plug misfire could lead to the damaging of the catalytic converter.
GM also mentioned that light-duty truck operation produces more
ignition events in 30,000 miles than does a passenger car because
of the generally higher N/V ratios. No commenter believed that
30,000 mile spark plugs were feasible over the entire useful
vehicle life. During the first 30,000 miles, the condition of the
engine is excellent and deterioration is neligible. However, as
mileage is accumulated wear takes place which in turn could cause
shorter acceptable spark plug life intervals.
All commenters agreed that they are unaware of the required
technology to guarantee the 100,000 mile maintenance interval for
catalytic converters in LDTs. GM proposes that a lesser interval
of 75,000 miles would be more logical considering the extended
vehicle life of 130,000 miles. Catalyst replacement at 75,000
miles will provide a fresh catalyst much earlier in the service
life. Several commenters believe that EPA has not established the
need or the feasibility of the proposed 100,000 mile requirement
for catalysts. They contend that EPA data on two passenger cars
(reported in MSAPC-79-211-B-4) are inappropriate for LDTs due to
the more severe service environment. GM mentions that even in less
severe passenger car service occasional replacement of catalysts is
required in order to comply with the 50,000 mile requirement. The
U.S. Department of Commerce also took issue with the 100,000 mile
catalytic converter replacement interval as being unrealistic. DOC
believes that the LDT environment is more demanding and that
technology does not exist for 100K mile catalysts in passenger
cars. Therefore, theyi say, it is unreasonable to believe LDTs can
meet the 100,000 requirement.
GM comments that in their investigations on improving cataly-
tic converter life, they discovered that oil off the cylinder walls
had a greater degree of phosphorus toxicity for a catalyst than oil
that comes .down through the 'intake valves. They stated that
reduction in oil consumption is being worked on with the final goal
being longer converter life. GM also states that catalyst deterio-
ration evidence shows exponential decay, so catalyst decay is
expected to be less in the second 50,000 miles than in the first
50,000 miles of vehicle operation.
Comments were also received on^other maintenance intervals
proposed by EPA. Many manufacturers, and the U.S. Department of
-------
Commerce, criticized the minumum life requirements of oxygen
sensors. GM particularly has concern for the absence of mainten-
ance on the EGR system over the 130,000 mile useful life of LDTs.
IHC and Cummins questioned the maintenance intervals for turbo-
chargers and injector tips. Cummins believes the maintenance
interval for turbochargers on LDTs should be investigated and may
be different than turbocharger maintenance intervals for heavy-duty
engines.
The final area of comment is directed at the four satisfaction
criteria which were proposed to assure maintenance performance in
the field. Criteria (A), (C), and (D) were criticized for vague or
confusing language and criteria (B) for being illegal.
If the only option available to a manfacturer is criterion
(B), then it would be required to pay for the maintenance. Both
Ford and GM suggests that such a requirement contradicts §207(g) of
the CAA by placing the maintenance burden on the manufacturer
rather than the owner.
Ford also comments on the aspects of two other criteria (in
addition to the use of vague terminology). Criterion (C), they
say, will not be applicable to a situation where the only change in
the recommended maintenance is to adjust the interval. Also, Ford
reads criterion (D) to mean that when a signal is used to encourage
maintenance performance, the signal must be removed after survey
data has been collected. The data is of "doubtful utility" in such
a case.
GM commented that the options other than the manufacturers
paying for maintenance are illusory. Regarding criterion (A), GM
felt the purpose of scheduled maintenance is to prevent the deter-
ioration of the part to the point that irreparable harm is done to
other components and to insure this result," the maintenance is
scheduled with an adequate safety margin. Based on the safety
margin, it is very unlikely that the part will fail "precipitously"
at the proposed interval.
GM went on to criticize criterion (C) claiming that even if a
manufacturer could demonstrate that enough service parts are sold
each year, there is no way that maintenance performance can be
demonstrated in view of the multi-faceted aftermarket industry and
do-it-yourself maintenance. Additionally, such maintenance data is
simply not available and is not likely to be available.
i
Criterion (D) is criticized by GM as being illogical in that
there would be no possibility of obtaining the necessary data for
seven to ten years. GM further pointed out that the instrument
panel is space limited for the placement of such a visual signal
and only minimum necessary information should be required.
2. Analysis of the Comments
This section presents the EPA staff's di sens si on nr>_4 analysis
10
-------
of the comments summarized above. The comments will be treated in
the same order that they appear in the Summary of Comments. Since
many of the comments are the same as comments received on the
recent heavy-duty rulemaking proceedings, the reader is referred to
the analysis done in support of that action as a further refer-
ence.^/ This section begins with an overview of EPA's position on
allowable maintenance in general to provide a context for the
discussion.
By restricting the amount of emission-related maintenance
allowable during durability testing, EPA is primarily trying to
encourage an effort on the part of the manufacturers to reduce the
amount of owner attention that their emission systems require.
This encouragement fits into the larger strategy of sustaining the
air quality benefits of regulatory actions as the vehicles/engines
are actually used. Indeed, both the U.S. General Accounting Office
and the Automobile Association of America have recently pointed to
increased light-duty vehicle emiss'ion system durability as an
approach to better in-use emission performance in those vehicles.4/
5/
Certainly a functioning network of inspection-and-maintenance
programs will help achieve proper maintenance in the field, but
a complete network does not yet exist for light-duty trucks.
Likewise, the providing of clear maintenance instructions to the
user will also help to some extent. Again, this in itself is not a
total solution because the nature of emission control systems is
often such that the operator is not aware that maintenance is due
or that it is necessary. Thus, manufacturers have a real oppor-
tunity to help ensure in-use emission-system performance by pur-
suing long-lived designs that require little attention. EPA
expects that once resources are directed toward these design goals,
manufacturers will be able to reduce required maintenance well
below that necessary for current technology components.
Section 206(d) of the 1977 Clean Air Act Amendments (CAA)
directs that "[t]he Administrator shall by regulation establish
methods and procedures for making tests under this section," (i.e.,
tests to determine emission compliance). It is on the basis of
Section 206 that EPA's entire certification and durability programs
have been built, as well as the Selective Enforcement Auditing
regulations.
The commenters show concern that there is no specific Congres-
sional mandate for EPA to establish minimum technologically feas-
ible maintenance intervals for durability test engines. However,
the proposed maintenance requirements easily fall within the rather
broad wording of §206. Even certification and durability testing
as they appear in present regulations are not specifically de-
scribed in §206, yet they have never been successfully chal-
lenged. The requirement for the design of a certification program
is that vehicles and engines be tested "in such a manner as
[the Administrator] deems appropriate"^ Sect ion 206(a)(D). The
It
-------
"appropriateness'1 of the proposed changes is discussed later in the
context of the "factual basis" comments.
Among the responsibilities of the Agency under §207(c)(3),
is to make certain that the maintenance instructions provided to
owners require no more maintenance than necessary to assure emis-
sion compliance. A manufacturer should not be allowed to avoid its
warranty obligations by requiring excessive maintenance that is not
performed widely in the field. This would result in the voiding of
many warranties because of a failure to properly perform the
maintenance even though such maintenance was not actually necessary
to keep the vehicle or engine in compliance. Therefore, except
under adverse driving conditions, the maintenance required of
the owners to retain their warranty should not be more than that
performed during the certification testing. The conclusion, then,
is that the maintenance instructions should be based on the main-
tenance done during §206 durability testing.
Addressing specifically Ford's comment that the scheme for
maintenance introduces an "impermissible degree of uncertainty into
the emission certification process," the staff feels that the
comment is justified. Therefore, the staff recommends a provision
be included specifying a sufficient notification, similar to
86.084-22 (e)(l) (ii), to reduce the degree of uncertainty on new
or adjusted maintenance intervals of "emission related" com-
ponents. The new provision would allow the Administrator to
specify new or adjusted emission related maintenance intervals only
if the Administrator has previously notified the manufacturer of
the said maintenance interval no later than September 1 of the
given calander year two years prior to the model year. Provisions
for an appeal process should the manufacturers disagree with EPA's
decisions should also be included.
The logical and factual basis for establishing technologically
feasible maintenance intervals was challenged from several direc-
tions, but little information to substantiate the claims was
provided. We are not convinced that the degree of maintenance
required to maintain emission compliance is widely performed,
especially when component designs require frequent attention and
when performance of the maintenance does not improve driveability
or fuel economy.
Clearly at the emission levels proposed in this package,
proper maintenance is a key part of an overall in-use emissions
control plan. The weakness of the present regulations is the lack
of incentives for the required maintenance to actually get done.
The regulations here being challenged address one facet of the
problem by encouraging all manufacturers to use the best technology
components possible from a low-maintenance requirement standpoint.
The staff views the argument regarding market pressures on
component durabilities to be somewhat misdirected. If lower
maintenance in some components indeed provides a powerful competi-
10.
-------
tive advantage, then the market should be an important factor in
designs beyond today's technology. Generally, however, we do not
believe that the market pressures for improved durability in
emission-related components is strong. (The durability of emission
controls has not been widely stressed in advertising, for example.)
The staff is also concerned with the implication that manufacturers
would be willing to trade off improved maintenance characteristics
and durability (and hence, a degree of better maintenance in the
field) for commercial purposes. We cannot accept the argument of
the existence of market pressures as a rationale for allowing more
frequent maintenance than present technology has been shown to
require. Conversely we do hope that the pressures will, in the
future be a strong factor in encouraging continuing reductions in
the amount of maintenance required on emission-related components.
The issue of spark plug maintenance intervals will now be
addressed. A majority of the manufacturers currently recommend a
maintenance interval range from 22,500 miles to 30,000 miles for
light-duty trucks using unleaded fuel. Ford specifies a 30,000
mile spark plug maintenance interval for all of their 1980 light-
duty trucks. The 1984 heavy-duty vehicle regulations specify a
technologically necessary spark plug replacement interval of 25,000
miles. This interval came as the result of a fairly extensive
analysis of heavy-duty engine operating conditions and current
intervals.3/ Light-duty truck application is less severe than
heavy-duty application. Thus, the minimum necessary interval for
light-duty trucks will be greater than 25,000 miles. The current
use of a 30,000 mile interval by Ford argues strongly in favor of
that number.
Commenters criticized the spark plug replacement interval in
two major areas: higher N/V ratios when compared to light-duty
vehicles, and increased oil consumption on vehicles with high
mileage causing spark plug deterioration.
There is no justification to reduce the spark plug replacement
interval because of the higher N/V ratios of light-duty trucks over
light-duty vehicles. The manufacturers' recommended interval
ranges of 22,500 miles to 30,000 miles for light-duty trucks is the
same as for light-duty vehicles. Clearly, if higher N/V ratios of
light-duty trucks over light-duty vehicles are a concern, the
manufacturer would have specified a lesser, different interval than
those specified for light-duty vehicles.
G.M. and IHC point out that light-duty trucks with high
mileage tend to burn more oil and cause increased spark plug
deterioration. This comment is unjustified when reviewing the
current manufacturer recommended spark plug replacement interval.
Current practice in the industry does not recognize high mileage as
a factor calling for reduced spark plug intervals. While the
manufacturers recommend a reduced maintenance interval for the
checking, cleaning and regaping of spark plugs under severe opera-
ting conditions, the manufacturers' maintenance schedules do not
-------
specify a reduced replacement interval on spark plugs as vehicle
mileage increases, but indicate only one spark plug replacement
interval. Additionally, G.M., in another area of comment, states
"[T]he maintenance is scheduled with an adequate safety margin to
accommodate some portion of the variation in part and in the
manufacturer assigns "an adequate safety margin" in the replacement
interval and does not indicate a replacement interval reduction
with increased vehicle mileage in their current maintenance sche-
dule, then the manufacturer's concern about increased oil consump-
tion causing increased spark plug deterioration must be overstated.
Considering the spark plug interval for the 1984 heavy-duty
regulations and the arguments given above, the staff does not see
any need to reduce the proposed 30,000 mile spark plug replacement
interval.
We now turn to an analysis of the comments relating to the
proposed 100,000 mile catalyst replacement interval. The comments
generally took issue with EPA's extrapolation of light-duty vehicle
catalyst technology to light-duty truck application and the lack of
data supporting the proposed interval. However, not one manufac-
turer supplied data supporting arguments that 100,000 mile replace-
ment intervals are unrealistic and cannot be achieved.
The staff rejects the contention that light-duty trucks are
significantly different than light-duty vehicles. The manufac-
turers, with exception of IHC, use engines and catalysts that are
the same as in light-duty vehicle installations. Obviously, the
manufacturers cannot claim a significant difference between light-
duty vehicles and light-duty trucks and yet use identical catalysts
for both applications.
To determine catalyst replacement intervals, a best estimate
must be obtained from the information available. The methodology
used to determine the interval is well documented in the heavy-duty
vehicle analysis.^/ A review of the specified heavy-duty vehicle
catalyst replacement interval shows that the light-duty truck
interval is conservative. Additionally, AMC recently provided
100,000 mile certification data on four light-duty vehicles that
substantiated the proposed interval is technologically feasible.
Three out of the four vehicles indicated no catalyst failure over
100,000 miles. The catalysts used on these AMC vehicles are
considered to be lightly loaded, i.e., a low amount of catalyst
material per unit volume. Lightly-loaded catalysts are considered
to be less durable than the heavy-loaded catalysts which may be
used on light-duty trucks. Clearly, if the technology exists to
produce a lightly loaded catalysts that lasts for 100,000 miles, a
catalyst design for light-duty truck application capable of lasting
100,000 miles is also feasible.
The staff disagrees with GM's recommendation that a lesser
replacement interval of 75,000 miles be used on catalysts. The
75,000-mile replacement interval would happen after 60 percent of
-------
the new anticipated "useful life" of 120,000 miles occurs. Clear-
ly, the replacement interval should be placed as close to the
vehicles' "useful life" as technologically feasible in order to
accomplish EPA's intent of reducing the amount of owner attention
to emission systems and sustaining the air quality benefits
as vehicles are actually used. It is doubtful the recommended
75,000-mile interval catalyst change with 60 percent of the vehic-
le's useful life over would be done by the owner. By specifying
the replacement interval, i.e., the 100,000-mile interval, close to
the anticipated useful life, the staff's concerns that the catalyst
may not be replaced are minimal. In fact, it is most likely that
such a well designed catalyst would perform for the full useful
life without any replacement.
In addition to the spark plug and catalyst interval, com-
menters showed concern over the absence of maintenance on the EGR
system for the useful life of light-duty trucks. The absence of
an EGR maintenance interval was accidental. A 50,000 mile main-
tenance interval will be placed on the EGR valve. This interval is
supported by previous studies and is the same interval specified in
the 1984 heavy-duty vehicle regulations.^/
The minimum replacement interval of oxygen sensors was criti-
cized by the U.S. Department of Commerce and many manufacturers,
but data arguing against the 50,000 mile interval were not pre-
sented. To the contrary, Ford specifies a 50,000 mile replacement
interval for their oxygen sensor used on light-duty vehicle appli-
cations and therefore demonstrates that a 50,000 mile oxygen sensor
is technologically feasible for light-duty vehicles. Additionally,
the 1984 replacement interval for heavy-duty vehicle oxygen sensor
as set by the 1984 heavy-duty vehicle regulations is 50,000 miles.
Light-duty- truck application is considered to be less severe than
heavy-duty vehicle application. With the technologically feasible
replacement interval for oxygen sensors of 50,000 miles established
for both light-duty/heavy-duty vehicles, the need to alter the
50,000 mile replacement interval for light-duty trucks can not be
justified.
The final area of comment regarding maintenance intervals is
specific to diesel engines and was directed to the maintenance
intervals of the turbochargers and injector tips. Cummins com-
mented that the maintenance intervals may need to be different than
the intervals specified for heavy-duty vehicles. The staff recog-
nizes that light-duty truck engines have shorter lifetimes than
heavy-duty engines and therefore will reduce the maintenance
intervals from 200,000 miles to 100,000 miles for both turbo-
chargers and injector tips.
We recommend that EPA delay the requirement that manufacturers
must demonstrate "a reasonable likelihood" that proper maintenance
will be performed in-use. Our recommendation arises not from
specific comments about these proposed provisions but from a belief
that such a requirement is not necessaty at this time. It appears
-------
to us that the manufacturers would reasonably easily be able to
show that required maintenance was indeed being performed on the
emission-related components which these .regulations will require.
With respect to the forthcoming NOx regulations, however, the
situation is different. It is possible that three-way catalyst
technology will be used, in which case oxygen sensors will control
the feedback systems. It is for this type of component that the
staff believes some sort of assurance of in-use maintenance will be
necessary. Regarding catalysts, assuming they cost more than 2
percent of the vehicle cost to replace, then the manufacturer is
required to bear the cost of replacement with or without these
provisions.
4. Summary of Recommendations
The staff has concluded that the proposed maintenance require-
ments (Section 86.083-25 of Subpart A) should be retained in their
proposed form with the following exceptions:
a. A 50,000-mile EGR system maintenance interval should be
included.
b. Include a provision to specify a. sufficient notification
to the manufacturer on new or adjusted maintenance intervals of
emission related components.
c. Delete the requirement that manufacturers must demon-
strate "a reasonable likelihood" that proper maintenance will
performed in-use.
d. Change the diesel engine turbocharger and injector
maintenance intervals to 100,000 miles.
-------
References
\J 1983 and Later Model Year Heavy-Duty Engines, Proposed
Gaseous Emission Regulation," 44 FR 9494, February 13,
1979.
2l "Paccar, Inc. V. National Highway Traffic Safety Administra-
tion, 573 F.2d 632 (1978).
3/ "Summary and Analysis of Comments to the NPRM 1983 and Later
Model Year Heavy-Duty Engines Proposed Gaseous Emission
Regulation," EPA, OMSAPC, December, 1979.
4/ "Better Enforcement of Car Emission Standards — A Way to
Improve Air Quality," Report by the Comptroller General of the
U.S. General Accounting Office, Report IICED-78-180, January
23, 1979.
5/ American Automobile Association letter to Rep. Henry Waxman,
~~ August 13, 1979 (EPA Central Docket Section #OMSAPC-78-4).
6/ See Issue G - Technological Feasibility.
7f "Control of Air Pollution from New Motor Vehicles and Motor
Vehicle Engines: Gaseous Emission Regulations for 1984 and
Later Model Year Heavy-Duty Engines, 45 FR 4136, January 21,
1980.
8J "Emission-Related Maintenance Intervals for Light-Duty Trucks
~ and Heavy-Duty Engines," R. A. Rykowski, EPA, OMSAPC, Tech-
nical Report No. SDSB-79-09, January, 1979.
-------
D. Issue: Idle Test and Standards
1. Summary of the Issue
EPA has proposed separate certification standards and test
procedures for the idle mode for both gasoline and diesel light-
duty trucks.
2. Summary of Comments
The manufacturers agree unanimously that both the proposed
idle test and the idle standard are redundant since the Federal
Test Procedure (FTP) currently includes the idle mode as ap-
proximately 18 percent of the transient cycle. Furthermore, most
manufacturers agree that their vehicles, since they comply with
FTP, will inherently meet the idle standards and that EPA should
therefore withdraw both the idle test and the idle standard as
applied to light-duty trucks.
Several manufacturers identified the proposed idle test as
confusing and insufficiently defined. Another major concern
was that poor correlation between the FTP and the idle test
would result in the requirement that manufacturers certify vehicles
to two different sets of standards.
The intent of EPA in requiring the idle test/standard was
also questioned. One manufacturer indicated that if the stan-
dard was designed to evade test correlation requirements as re-
quired under section 207(b) it was, "against the intent of the
Clean Air Act." If designed for Inspection/Maintenance (I/M)
purposes, the question was raised concerning the development of a
specific I/M test that would not require adjustment of the cer-
tification process; an adjustment that would, according to the
commenters result in increased certification costs.
The MVMA identified two major concerns with regard to the
"supposed desirability" of separate idle requirements. First,
was the claimed lack of control over testing conditions, equipment
calibration and personnel training which would, in their opinion,
result in questionable test validity. Secondly, according to the
Association, any short test that does not measure all regulated
parameters, ''could not be employed to invoke the warranty require-
ments of Section 207(b)," since the proposed test does not measure
all regulated emissions.
Additional comments include that:
a. EPA has not established that CO "Hot Spots" exist (jus-
tification is a requirement if these "Hot Spots" are the basis for
the idle test).
b. EPA has not established that a 90 percent reduction in HC
and CO during idle is required to improve air quality.
18
-------
c. EPA has not established that HC analysis is required for
the idle test.
d. EPA has not established that a separate idle test is a
necessity for diesel vehicles.
3. Analysis of Comments
Based upon available data the staff agrees with the manufac-
turers that catalyst equipped vehicles have the capacity to meet
the idle standards as applied to light-duty trucks. The staff does
not, however, view the idle standard/idle test as redundant, but
rather as a method of assurance that the available technology is
actually controlling idle emissions. The idle test serves as a
design criteria to insure against any tradeoff of idle vs non-idle
emissions. This criteria, we believe can be met easily and for
little or no cost.
The current FTP for light-duty trucks was designed with
measurement of hydrocarbons as the prime consideration. By its
nature as a representative driving cycle, such a procedure would
also be an accurate measure of other pollutants being emitted
(such as CO). However, at the same time, it may not be the optimum
procedure for all pollutants in terms of their environmental
impact, because different pollutants have different mechanisms of
operation. HC emissions combine on a fairly broad scale, and over
a period of a few hours participate via photochemical reactions in
the formation of ozone. CO on the other hand, is toxic in its own
right. Highest concentrations are typically associated with
morning and afternoon rush hour traffic, and can be highly local-
ized. The driving conditions associated with central business
district rush hour traffic would have a lower average speed and
higher amount of idle than does the FTP. Also, at very low speeds,
tailpipe emissions can be substantially affected by carburetor idle
settings. The idle test for CO contained in this rulemaking is a.
direct way of insuring that reductions in CO seen on the FTP will
also occur during low speed, high idle fraction operation chara-
cteristic of urban rush hour driving.
i
Furthermore, costs of compliance with the certification idle
test are minimal. As discussed above, catalysts effective on the
transient certification procedure should easily meet the certifi-
cation idle standard as a matter of course, therefore, requiring no
additional development costs. The only attributable costs to the
idle test procedure are those associated with performance of the
actual test for certification, which is insignificant on a per
engine basis.
Diesel engines, however, will not be equipped with emissions
sensitive catalyst systems. Use of an idle test procedure for
fi
-------
diesels, with or without in-use compliance testing, is expected to
have little or no effect.
The staff also agrees that the necessity for HC analysis in
the idle test has not been established. Therefore, it is recom-
mended that the HC idle standard be eliminated.
The concern that was expressed that EPA is evading the 207(b)
requirements should be dispelled by the fact that warranty regula-
tions and the associated short test procedure have been recently
adopted in a separate rulemaking (45 FR 34802). We do not rely on
§207(b) for authority for this idle test procedure.
A number of specific technical issues on details of the idle
test procedure were raised during both this comment period and
the earlier comment period on the heavy-duty gaseous emission
regulations. As a result, a number of changes are suggested
for the final version of the test procedure of Subpart P:
a. The idle HC standard for gasoline engines will be de-
leted.
b. The limitations called for in 86.1511-83(a)(l)(i) and
(ii) will be adjusted from 45 to 90 percent, to 15 to 100 percent.
c. The interference gas called for in 86.1511-83(a)(7)
should be N0£ at a 100 ppm concentration not NOx at a 1000 ppm.
Diesel comments do not apply here since Subpart P will apply only
to gasoline-fueled engines and vehicles.
d. A temperature controlled environment as referred to in
86.1511.83(b) is defined as the interior of a normally heated
room.
e. Section 86.1514-83(a)(2) is adjusted by the heavy-duty
procedure.
f. Utilization of the word "calibration" in section 86-
1516(b) is incorrect; "calibration " should be replaced by "span."
g. The average ambient temperature of the vehicle environ-
ment should be changed from -20CC to 45°C (-4°F to 113°F), to 2°C
to 43°C (35°F to 110'F).
h. Identification of operators in section 86.1542-83(a)(5)
and (6) will not be deleted.
-------
i. Additive deterioration factors (DFs), are to be sub-
stituted for multiplicative DFs in the case of the idle test. This
adjustment is being made due to nondetectable or very low emission
levels during LDT idles.
In summary, the high percentage of time that all vehicles
spend at idle, the ease of an in-use idle procedure, the improved
effectiveness of an in-use idle procedure in detecting failed
catalyst systems, and the virtually nonexistent costs of a cert-
ification idle standard support its promulgation. No compelling
data at this time, however, support implementation of any idle
standard for diesel engines, and a delay in its promulgation is
warranted. This decision could be reconsidered in the future,
should the need become mord evident.
4. Recommendations
Retain the idle CO standard for gasoline engines. Delete the
idle test requirements for diesel engines; delete the idle HC
standard for gasoline engines.
-------
E. Issue: Leadtime
1. Summary of the Issue
In the NPRM, EPA proposed to implement the new light-duty
truck regulations for the 1983 model year. This was based upon
EPA's belief at that time "that sufficient leadtime is available to
develop and apply the necessary emission control technology and to
conduct compliance testing by 1983." (44 FR 40792, July 12,
1979). The NPRM went on to say, "(t)herefore, EPA cannot make the
findings under Section 202(a)(3)(C) necessary to permit consider-
ation of revised standards at this time." At the same time,
however, manufacturers comments on the feasibility of compliance
with the proposal in 1983 were solicited. The request for such
comments stated that "in order that EPA may fairly evaluate the
technical merit of all comments on feasibility, the Agency requests
that comments be accompanied by supporting data or other informa-
tion."
2. Summary of the Comments
EPA received a number of comments on the feasibility of
complying with the proposed 1983 deadline. These comments fall
broadly into two areas: (a) technical feasibility, and (b) legal
requirements of the 1977 Clean Air Act Amendments.
a. Technical Feasibility of the 1983 Model Year
Comments on the technical feasibility identified the proposed
standards as achievable for 1983 if viewed in the context of
currently existing certification procedures. The additional
aspects of the proposal dealing with allowable maintenance, in-use
durability, redefined useful life, and SEA were seen by commenters
as making the available leadtime marginally or wholly infeasible.
An example of this position is the following comment by GM:
"General Motors is concerned that there is inadequate leadtime
to asure compliance with the entire regulatory package. While
we do not foresee any insurmountable leadtime obstacles in
incorporating" the "best effort" hardware described in Section
IV in time to meet start of production for 1983 MY light-duty
trucks, this does not imply that there is adequate leadtime
allowed. If this "best effort" does not succeed in meeting
the regulations as finally adopted, further changes would have
to be made. At this point in time, we do not know what this
would entail. It may include further hardware changes,
revisions to production techniques, further improvements in
quality control procedures, or other things we are not now
aware of. Until we have experience with the "best effort"
system under the proposed 1983 certfication procedures, we do
not "know"—we can only speculate ."_!/
Comments of the above sort expressed a. lack of information
-------
from which to determine whether currently envisioned emission
control system changes would be capable of meeting the overall
requirements of the proposal. These comments are set forth in
greater detail in the analysis of feasibility elsewhere in this
Summary and Analysis of Comments.
IH and Ford developed and submitted specific timelines for
their compliance programs. No other manufacturers provided such
timetables.
The IH schedule, reproduced here as Figure E-l, led IH to
conclude that "such timing greatly jeopardizes a manufacturer's
chances of ensuring a certified produceable product line."2/ The
major time elements of the IH timetable are devoted to establishing
full life deterioration rates. IH projected a need for a two phase
testing program to establish durability data and design systems for
extended life. Each phase involves the generation of high mileage
data from a fleet of vehicles. The first phase is intended by IH
to establish basic system requirements. The second phase would
involve a fleet of production prototype vehicles run for the full
usefull life to determine preliminary DF's. Together, these two
phases would require two years to complete.
The schedule as developed by IH was predicated upon promulga-
tion of the final rulemaking by January 1, 1980. It contained a
four month interval between the two phases of the durability
program for supplemental development or redesign required as a
result of the first phase testing. That four month interval was
"not considered feasible" to accomplish the necessary work.
Figure E-2 presents the timing plan developed by Ford. As
with IH, the largest single element of the timetable is the program
for development of durability data (labeled "useful life emission
development"). Ford projects that this would take approximately
two and a half years to complete. The first six months are con-
sumed in assembling the prototype vehicles to be used in the
durability fleet (six per engine family). This is followed by
mileage accumulation on a staggered basis to reduce the testing
facility load. Comple.tion of mileage accummulation would then
occur over a six month timespan from the first to the last vehicle.
Ford indicates that procurement of prototype durability vehicles
would have had to begin by May 14, 1979 to allow time for certifi-
cation to the 1983 model years.
American Motors and IH both argued that they were vendor
dependent for development of new control systems. They indicated
that their corporations lacked the resources needed to develop
these systems for themselves. AM stated that "the overall regu-
lation would be unachievable even with the most advanced concepts
presently under development of our passenger cars."5/ IH indicated
their suppliers might be their own competitors.
Both manufacturers indicated that bethg vendor dependent led
20
-------
Figure E-l
IH LEAD TIME SCHEDULES/
8/01/79
8/01/80
8/01/81
11/01
2/01
5/01
11/01 2/01
12/01 . A/01
8/01/82
5/01
IUMINGS
•83 STDS
-FINAL
RULES?
100K MILES PRE-
IHMiLOPMnNT DURABILITY
13 MQMH DURADILI1Y VHIICLnS
DATA
VEHICLES
45
DAYS
ALL PAPER
TO DPA
FOR CiiRT.
-------
LIGHT TRUCK NPRM L JFUL LIFE TIMING PLAN
1979 1980 1981 1982 1
A MJJJJ AJS O;N|O jiF'MJAlMjjj j|A s|o N D J FMAMJJASOMDJ FMAMJJASONDJFMAM
; -.''v » i: •-••! • •> I ••-•-. '-•••-••'•'• '• •" - '-' 1J 1' '<"• '••' '•' '•' '•' '• '0 1 8 7 ij ;.|4 3 :• i -1 2 -3 4 -5 -6-789 -10
Slid Slut Fisl Compiee XI
PV.IO O'x: frsi !esi lesi tea / 1
4-14-80 10 1380 5-15-81 11-9-61 / 1
T » » T / JOb ffl
®j USEFUL LIFE EMISSION DEVELOPMENT | <\ 8-i-"ii?
•.147!) tll'I.N ' ijj,, '' <;,„, ' COMHIltt CtfU N. 1
'•|J" . '4lll|l 1 IK ton ric'vuk(i leil [PA CAflO ^
Ii. in i1"*. I«M 1.-,: t ij.,,, 9.7.8, 2.! ].|.B2
r MISSION or /i locuw @| 4K TEST & CERTIFICATION 1 A^^aSi
I.... ._
IN-USE USEFUL
50X fUISSIOM DEVUOPMCHT n»iY*ill
lillll Slnfl I'll CornJuU
Piolo Pioc. Fril IK) TKI Tnl
(-13-81 101781 5148? 11882
T » » T
® 1 • USEFUL LIFE EMISSION DEVELOPMENT ]
?. \-tw ' ii 1780 su" :-l'»l [nuun Dove** lot! II'A CAIU
1*1*1 hi. Irn lin.1 J-15-82 9-6-82 2-1 3-1-83 «
t 'if Mhi I) u'jii ui uff Y * • .- T_. T
IMI- .-.,.-» imn.M-MiH. (D| 4K TEST a CERTIFICATION |
50K tMlSSIDM CEVEIOPMFH! Com*>ii'
11, .IU I1l«. ll>l lOkl IlILl
4 cn;1 in u ii.' 'i u HI
V T »
@f USEFUL LIFE EMISSION DEVELOPMENT
if- ------ 4 '• • • •
'j 10 UI II iSBl sum
Su«l Sim Cmuxi Develop
FtoloPioc litl Inl 2-14B3
983 1984 1985
JJASONDJFMJAMJJASONDJFMAMJ
1 -12 lo 14-16-HJ-I7-U) 19 ^o|-2H>2-2.)
-------
to longer leadtime requirements than would otherwise be the case.
This they felt placed them at a significant disadvantage. AM
stated that they would need "at least two extra years of lead-
time...from the time such demonstrated components are made avail-
able to us from our suppliers."
Lastly, the Department of Commerce expressed the belief that
"(t)he advanced engineering phase of the manufacturers' 1983
light-duty truck program is now well underway with production
engineering scheduled to start within the next few months. There-
fore, it is essential to establish regulations to meet the 90
percent reduction objective that introduce a minimum of uncer-
tainty. %/
b. Legal Requirements of the 1977 Clean Air Act Amendments
Commenters expressed the belief that in the 1977 Clean Air Act
Amendments, Congress mandated a minimum four year leadtime appli-
cable to those light-duty trucks which fall within the statutory
heavy-duty truck class. (The statutory heavy-duty truck class
includes light-duty trucks in the 6,000-8,500 Ibs GVW class.) As
stated by MVMA, "the 1977 Amendments and the accompanying legisla-
tive history unequivocally demonstrate that Congress intended to
require EPA to promulgate 1983 standards so as to provide at least
four years leadtime to manufacturers. There is not a single
statutory provision or element of legislative history to the
contrary." They stated that other provisions of the Amendments
(such as those for temporary or permanent revisions under Sections
202(a)(3)(B) and 202(a)(3)(E)) "indicate that Congress in fact
intended to override the discretion as to leadtime otherwise
accorded to EPA by Section 202(a)(2)."7/
To support this position, commenters cited provisions of the
Act, and items of legislative history. Section 202(a)(3)(B) states
that "(d)uring the period of June 1 through December 31, 1978, in
the case of hydrocarbons and carbon monoxide,...the Administrator
may, after notice and opportunity for a public hearing promulgate
regulations revising any standard prescribed as provided in sub-
paragraph (A)(ii) for any class or category of heavy-duty vehicles
or engines. Such standard shall apply only for the period of three
model years beginning four model years after the model year in
which such revised standard is promulgated." The standards of
subparagraph (A)(ii) referred to here are the statutory 90 percent
reductions applicable to 1983 and later model years. These pro-
visions are taken by commenters to mean that EPA was to have
finalized by the end of 1978 those standards which manufacturers
would be required to meet in 1983 - thus allowing four years of
leadtime for compliance. Commenters believed that this timetable
was to be followed by EPA even if the revision procedures of
Section 202(a)(3) (B) were not being invoked. If EPA was required
to allow four years for a relaxed (less stringent) standard, they
argued, then it would be illogical for EPA not to have to provide
-------
four years for the more stringent statutory reductions to be
implemented.
As further support for the Congressional intent argument, the
following excerpt from Section 202(a)(3)(E)(ii) (which provides for
"permanent" changes in the statutory standards) was cited: "No such
changed standard shall apply for any model year before the model
year four years after the model year during which regulations
containing such changed standard are promulgated." This citation
is considered a further example of what commenters identified as
clear Congressional intent for provision of four years leadtime.
Commenters also felt that if a choice had to be made between
four years of leadtime and implementation for 1983, that the four
years requirement must be met. As stated by the Department of
Commerce, four years "is the time considered by industry and this
Department as the minimum in which compliance can be obtained
without risking undue market disruption."8/
MVMA cited testimony provided before the House and Senate by
both heavy-duty manufacturers and EPA in 1977 as the Amendments
were being considered. The citation indicated that manufacturers
stressed, and EPA recognized, the need for sufficient leadtime.
MVMA also cited a statement by Senator Muskie in November of 1977
which indicated a conference agreement for four years leadtime. As
originally adopted, the 1977 Amendments had specified June 1
through December 31, 1979 as the period during which standards
could be revised. This was later amended to be June 1 through
December 31, 1978. Senator Huskies' explanation of this change was
that it was made "so as to conform to the conference agreement for
four years leadtime."9J MVMA acknowledged that the 1977 Amendments
do not expressly provide for promulgation of the statutory stan-
dards in 1978, but felt that this was the clear intent of Congress.
3. Analysis of the Comments
a. Technical Feasibility of the 1983 Model Year
Aside from IH and Ford, both of whom alleged specific timing
problems in meeting a' 1983 compliance date, commenters presented
more of a feasibility problem than a leadtime one. For example,
the comment from GM indicated sufficient time available to apply
hardware changes. The doubt expressed by GM was over the adequacy
of their "best effort" hardware to do the job. Other commenters
supported this position.
The issue of feasibility is addressed elsewhere in this
Summary and Analysis of Comments under "Technological Feasibility."
The analysis presented there demonstrates clearly the feasibility
of the final rulemaking in the form recommended by the technical
staff. Thus, the staff does not consider the comments of this sort
as affecting the viability of the 1983 model year, as proposed in
the NPRM, for implementation of the x£inal rules. However, the
-------
other leadtime questions raised by specific commenters do need
careful consideration.
1) International Harvester
The bulk of the time in the IH leadtime schedule (Figure E-l)
is devoted to mileage accumulation on durability vehicles. The
first phase (pre-development durability fleet) is scheduled to
require 11 months, and the second phase (production prototype
fleet) is scheduled to require 13 months. Both the need for two
phases of testing and the time required are questionable. The need
for this program was determined largely by the proposed in-use
durability requirements. IH felt that "the jeopardy inherent in
the in-use scheme mandates that every attempt be made to simulate
ahead of time, and as accurately as possible, the estimated final
DF's" In light of the staff recommendation to defer the in-use
durability program at this time, it is likely that IH would no
longer feel the need for a dual durability program. The staff
believes that even with an in-use program, the need for such two
phase testing was not demonstrated by IH. No other manufacturer
indicated an intent to duplicate their pre-certification durability
testing. The staff believes that the IH timetable should be
revised to eliminate the first phase testing.
In conducting its durability testing of prototype vehicles,
each manufacturer will be able to design the type of testing
program it feels is most appropriate. Because of the latitude
available to the manufacturer it is not possible to quantify
the exact amount of time this process will require. However,
the staff believes that IH would have no incentive to undertake a
testing program more rigorous or time consuming than the current
light-duty truck mileage accumulation procedure. Therefore, to
estimate the maximum amount of time to prove system durability and
establish deterioration factors the time to accumulate full
useful life mileage on the current light-duty truck durability
schedule will be used. A review of records on durability data
vehicle mileage accumulation from 1980 engine families indicates 3
1/2 - 5 months as typical times for accumulating 50,000 miles on
the EPA durability schedule. This time includes related aspects of
mileage accumulation such as maintenance and testing. Allowing the
maximum of 5 months would give an average accumulation rate of
10,000 miles per month. The average useful life of light-duty
trucks has been estimated by EPA as 120,000 miles.1Q_/ However, in
view of the uncertainty of this estimate, coupled with an allowable
catalyst change interval of 100,000 miles, the technical staff
believes that the useful life used by most manufacturers for
certification will fall at approximately 100,000 miles. If a
manufacturer indeed chose a lifetime longer than 100,000 miles, he
would still be able to end mileage accumulation at the catalyst
change .point. A new catalyst would substantially reduce exhaust
emissions.
At 10,000 miles per month, it would take 10 months to accu-
-------
mulate 100,000 miles. Allowing 1 month to prepare the durability
vehicle would bring the total time to 11 months. IH would have
sufficient time to conduct this program and development work in
time for the 1983 model year if the final rules are promulgated by
June 1, 1980. Development time would be restricted, however, if
the final rules are promulgated after June 1, 1980.
2) Ford
Turning now to the Ford leadtime schedule of Figure E-2,
leadtime for development of durability data is once again the major
factor. As with IH, Ford's extensive two and one half year dur-
ability program was principally motivated by the proposed in-use
durability program. Ford proposed testing a sizeable fleet of six
vehicles per engine family, staggering the mileage accumulation
over six months. With the deferral of the EPA in-use durability
fleet requirements, Ford would be expected to return to a mileage
accumulation program characteristic *of current practice, which,
as noted, could be completed in 11 months. Thus, Ford could
accomplish the desired 11/9/81 completion date of Figure E-2 by
starting durability testing before December 1, 1980 and also
have six months development time if the final rules are promul-
gated before June 1, 1980. Promulgation after June 1, 1980 would
restrict this development time.
3) Vendor Dependency
Both AM and IH claimed that their dependence upon vendors for
development and supply of new control systems posed a significant
leadtime problem. The EPA technical staff believes that these
claims are greatly exaggerated.
IH stated that "(c)ompliance with the proposed standards will
require technological improvements in the major areas of catalysts,
fuel system, air injection, and basic engine design and control
system parameters." Contrary to this position, the staff analysis
of feasibility (see Issue G: Technological Feasibility) indicates
that no major innovations are required. In fact, many 1980 Cali-
fornia light-duty truck models appear able to comply with the
recommended final rulemaking. Data submitted by AM for high
mileage catalyst equipped vehicles indicates the capability of
their current systems to perform for at least 100,000 miles.ll/
Vendor dependency may create some time penalty even though no
major emission control system changes appear to be required.
However, all manufacturers experience this problem to one degree or
another. For example, most manufacturers use vendor supplied
catalysts and must work with their suppliers when changes are
needed. While smaller manufacturers such as IH, AM and to some
extent possibly Chrysler, may be somewhat more vendor dependent
than larger manufacturers, such dependence is minor in consider-
ation of the available leadtime.
-------
The question of vendor dependency and supply leadtimes also
arises in connection with the Electronic Engine Controls (EECs)
which the staff is projecting to be used to meet these regulations
without incurring a fuel economy penalty. The staff forsees no
supply problem in adopting EECs as early as the 1983 model year.
In fact, in its recently proposed fuel economy standards for
light-duty trucks, NHTSA indicated its belief that EECs would be
available for the 1982 light-duty truck fleet—one model year
earlier than needed here. Respondents to NHTSA1s proposal did not
challenge the feasibility of that date.
4) Advanced Engineering
The Department of Commerce comment concerned the meshing of
EPA's final rulemaking requirements with the advance engineering
which DOC felt was already well underway, and with production
engineering which DOC felt was scheduled to begin "within the next
few months." EPA technical staff agrees that each manufacturer is
entitled to as much advance notice of requirements as is reasonably
possible. In evaluating the adequacy of timing in the present
case, the staff turns to comments of the affected manfacturers.
Apart from the DOC, there were no other comments on this issue. No
manufacturer felt the need to raise it as a concern from the
manufacturer's viewpoint. Therefore, the staff concludes that
this concern is not a valid issue in the context of the present
rulemaking.
b. Legal Requirements of the 1977 Clean Air Act Amendments
The technical analysis just completed in Section a above
indicates that compliance with the recommended final rulemak-
ing provisions can readily be attained in approximately two
years of leadtime before the model year of introduction. However,
we must now turn to an analysis of any legal constraints placed on
allowable leadtime by the Clean Air Act as amended in 1977 (the
Act). If, as maintained by commenters on the proposed rulemaking,
Congress mandated a four year minimum leadtime for those light-duty
trucks in the 6,000-8,500 Ib. GVW range, then the earliest model
year for which the final rulemaking could be applicable for those
vehicles would be 1985. The EPA staff believes that four years
leadtime is not mandated for those vehicle. Of course, vehicles
below 6,000 Ibs. GVW are not considered heavy-duty vehicles under
the statute, and as to them, there is no issue concerning a stat-
utorily prescribed amount of leadtime.
The comments received on this issue focused on the meaning of
several portions of Section 202(a)(3) of the Act. As indicated in
the Staff summary of these comments, the commenters believed that
such provisions must be interpreted as requiring four years of
leadtime for heavy-duty engines or vehicles. The staff rejects
this interpretation of the Act. EPA interprets the statute and its
legislative history as strictly requiring four years leadtime only
-------
in the case of revisions to the statutory standards. Although it
could be said Congress anticipated that standards under §202(a)(3)
(ii) would be promulgated at such a time as to be in place by the
last half of 1978, this assumption was never transformed into a
direct requirement. Rather, what was expressly required was that
the statutory standards apply to the 1983 model year.
The same issue has previously been raised in comments on EPA's
proposed gaseous emission regulations for heavy-duty engines (44 FR
9464, February 13, 1979). In promulgating the final regulations,
EPA delayed the year of implementation to 1984 based upon feasi-
bility considerations, thus making the legal issue moot. However,
EPA explained the relationship between the need for leadtime and
the 1983 implementation date as follows:
"It should be noted that it is the Agency's view that had it
been clearly evident that sufficient leadtime existed to
permit the manufacturers to comply with the statutory stan-
dards in 1983, Congress1 desire to have the standards take
effect in the 1983 model year would have taken precedence over
any expressions concerning leadtime to which the manufacturers
might, otherwise have been entitled to under Section 202." (45
FR 4144, January 21, 1980)
Indeed, this view is consistent with Congress' instructions
concerning the possible reduction of emissions in excess of 90
percent of baseline levels. See H.R. Rep. No. 95-294, 95th Cong.,
1st Sess. 273 n. 13 (1977). There, Congress recognized that if
technology were to be available to achieve reductions greater than
90 percent, EPA was expressly authorized to require its use.
Similarly, if technology will be available for implementation in
the 1983 model year, Congress would not have intended that its
implementation be delayed.
Clearly, EPA cannot satisfy both the 1983 implementation date
and provide four years leadtime in this rulemaking. At such a
point, the Agency must find the optimum policy for balancing
the environmental benefit versus the burden on manufacturers.
The goal which Congress surely had in mind in incorporating
leadtime guidelines into the act was to insure manufacturers
adequate time to respond to new standards without undue disruption
of their operations.
The four year requirement for revisions to the statutory
standards undoubtedly reflects consideration of the relatively
major changes in control technology facing manufacturers of
heavy-duty engines (over 8,500 Ib. GVW). These engines currently
use pre-catalyst technology and face the adaption of catalyst
systems into the heavy-duty environment. The technical analysis of
leadtime requirements for these manufacturers completed as part of
the final heavy-duty rulemaking concluded that four years would
indeed be required to accomplish those tasks.
-------
Those light-duty trucks which fall into the statutory heavy-
duty class (6,000 to 8,500 Ib. GVW) face a totally different
situation than that just described. Indeed, Congress recognized in
1977 that vehicles below 8,500 Ib. GVW present a somewhat different
set of regulatory problems and accordingly, expressly authorized
EPA to promulgate separate regulations for the class of light-duty
trucks. See Clean Air Act, Section 202(a)(3)(A)(iv); H.R. Rep. No.
95-564, 95th Cong., 1st Sess. 164 (1977) (Conference Report).
Even though most engines used in these light-duty trucks are also
used in heavy-duty vehicles, the light-duty truck applications are
more characteristic of light-duty vehicles than heavy-duty trucks.
Light-duty trucks are already employing oxidation catalyst systems,
and in some cases three-way catalyst systems. Their emissions are
measured using the same test procedure as light-duty vehicles. In
actual use they are treated more like cars than like heavy-duty
trucks, with most owners tending to view them as "big cars."
For all these reasons, the task facing light-duty truck
manufacturers in complying with the statutory 90 percent reductions
is much less difficult than that facing heavy-duty trucks. There-
fore, it is possible to provide adequate leadtime as Congress
desired without needing a full four years.
4. Recommendat ions
The EPA staff believes that in balancing Congress' desire to
implement the standards for 1983 with its desire to provide
four years of leadtime, the situation at hand dictates that
the 1983 deadline should take precedence providing that the
necessary technology and sufficient leadtime are available. The
issue of available technology is clearly presented under "Tech-
nological Feasibility" and has been demonstrated as feasible.
However, providing manufacturers with sufficient leadtime to comply
with the rulemaking is critical in establishing the model year
implementation date.
The EPA staff recommends retaining the 1983 model year
implementation date providing that the rulemaking is promulgated by
June 1, 1980. Recognizing that the promulgation of the rulemaking
after June 1, 1980 would place an increasing pressure on the
manufacturer to comply with the rulemaking, the EPA staff recom-
mends a 1984 model year implementation if the rulemaking is promul-
gated after December 1, 1980.
-------
References
_!_/ Comments of General Motors Corporation, October 11, 1979, pg.
89.
2] Comments of International Harvester, October 9, 1979, pg.
G-3.
JJ/ Comments of International Harvester, October 9, 1979, pg.
G-4.
_4/ Comments of Ford Motor Company, October 11, 1979, Attachment
III to Section IV.
5f Comments of American Motors, October 10, 1979, pg. 11.
6/ Memorandum of Comment of the Department of Commerce, pg.
17.
TJ Comments of Motor Vehicle Manufacturers' Association, October
11, 1979, pg. 68
8/ Memorandum of Comment of the Department of Commerce, pg.
~~ 14.
9/ 123 Cong. Rec. H 11957 (daily ed. November 1, 1977).
1Q/ "Average Lifetime Periods for Light-Duty Trucks and Heavy-Duty
Vehicles", EPA Report #SDSB 79-24, Glenn W. Passavant,
November 1979.
\\l Comments of American Motors, October 10, 1979, Appendix
B.
-------
F. Issue: Economic Impact
1. Summary of the Issue
The U.S. EPA has proposed a comprehensive control strategy for
1983 and later model year light-duty trucks.
This strategy includes more stringent HC and CO emission
standards, a new useful life definition, a revised durability
testing program, revised allowable maintenance provisions, and an
idle test with idle emission standards for HC and CO.
In addition, the control strategy includes a diesel crankcase
emission standard for light-duty trucks powered by diesel engines
and a reduction in the Selective Enforcement Auditing acceptable
quality level from 40 percent to 10 percent.
In the proposal the EPA technical staff estimated an average
per engine first cost increase of $62 (1978 dollars) with an
expected operating cost increase of about $60 (discounted to
January 1, 1983) associated with inspection/maintenance programs.
The rulemaking strategy as a whole was expected to cost approx-
imately $1.97 billion dollars for the 20.4 million light-duty
trucks sales projected for the first five years of this regula-
tion.
2. Summary of the Comments
The comments received on the economic impact of these proposed
regulations will be summarized according to the major components of
the rulemaking strategy. The following cost areas will be ad-
dressed: development and emission control hardware, certification,
allowable maintenance, useful life redefinition, diesel crankcase
control, and the stricter 10 percent AQL.
A. Development and Emission Control Hardware
Several manufacturers commented on the specific costs asso-
ciated with the emission control hardware and other developmental
costs associated with an emission control system capable of meeting
the emissions standards for the full useful life. These cost
comments are listed below by manufacturer.
1. Gasoline-Fueled LDTs
International Harvester
Exhaust System and Related Changes
Ignition System and Related Changes
Carburetion and Fuel System
Other
TOTAL $198
-------
Ford
Ford estimated costs at about $80 per vehicle, but this did
not cover the effects of full useful life or the 10 percent
AQL.
General Motors
General Motors estimated initial hardware costs of $362 per
vehicle. This figure includes $300 for the replacement of the
catalytic converter and exhaust pipe.
Chrysler
Chrysler estimated a customer cost of $50 per vehicle with a
50,000 mile useful life and tooling cost of about $500,000. In
addition, Chrysler stated that the full useful life might force the
use of three way catalysts and feedback carburetors. In this case,
the tooling cost was estimated at $20 million with a customer costs
of $315 per vehicle. Chrysler estimated emission control system
development costs at four million dollars.
Of the other manufacturers none commented on the emission
control hardware costs in sufficient detail for analysis.
2. Diesel LDTs
Of the three manufacturers which currently market diesel LDTs
none commented on the costs of bringing these engines into compli-
ance.
However, for the record, Chrysler provided costs for turbo-
charging light-duty trucks.
Turbocharger Unit - $ 243
Oil lines and other
plumbing charges - 27
Change in injection
pump configuration - 9
Manifold and turbocharger
exhaust transition - 46
TOTAL: - $ 325
B. Certification
No manufacturer commented in detail on the costs of recer-
tification. However, International Harvester commented on the need
for very complete testing due to the redefinition of useful life.
American Motors commented that EPA's certification costs were too
-------
low because too many ideal assumptions were made. American Motors
did not provide any revised cost estimates.
C. Allowable Maintenance Provisions
No specific comments were received on the costs of implemen-
ting the allowable maintenance provisions.
D. Useful Life Redefinition
The cost-related comments on the useful life redefinition can
be divided into two groups. The first group is greater costs aimed
at meeting the lower target levels associated with a longer useful
life. The second group is cost related to warranty claims.
1. Lower Target Levels
Chrysler
Chrysler speculated that 3-way catalysts and feedback car-
buretors might be necessary to meet the lower target levels for HC
and CO.
International Harvester
Although International Harvester did not comment explicitly on
the lower target levels, they did state that the more sophisticated
emission control systems used and the longer useful life period
anticipated would substantially reduce the value of their current
data on the durability of their emission control systems.
jFord
Ford Motor Company did not provide comment in sufficient
detail to allow analysis. They stated the extended durability and
revised durability testing procedures taken together would cost
$420. When questioned, Ford declined to provide further detail.
However, it appears that the bulk of the cost is tied up in further
hardware such as: electronic engine controls, feedback systems, and
additional light-off catalysts.
2. Warranty Costs
Several manufacturers provided comments on increased warranty
costs which they expected as a result of the longer useful life.
These would be costs related to Section 207 of the 1977 Clean Air
Act Amendments.
International Harvester
International Harvester provided in-depth comments on their
anticipated warranty related costs. Referring to the table found
in Attachment 4 to their written comments, IH's costs can be
-------
summarized as shown below:
1. Probable Warranty Cost (Parts and Labor) : $ 97
2. Diagnostic Costs (116% failure rate)
($23/hr x 2hr/diagnosis x 1.16 failure/vehicle) : $ 53
3. Dealer overhead and profit : $ 65
4. Average Engine Rebuild Cost
(100,000 - 130,000 miles) : $ 71
TOTAL : $286
American Motors
American Motors expressed serious concern over the warranty
implications of the extended useful life. They stated that the
full useful life would force each new truck customer to incur a
warranty related cost of $260 - $300 to cover emission related
warranty claims in the second half of the vehicle life.
Ford
No comments were received.
General Motors
No comments were received.
E. Diesel Crankcase Control
No comments related to the control of diesel crankcase emis-
sions were received.
F. 10 Percent Acceptable Quality Level (AQL)
In general, few comments were received on the costs of imple-
menting a 10 percent AQL. As a group, manufacturers felt that the
10 percent AQL, if feasible, would force lower target emission
levels and perhaps cause a fuel economy penalty.
Chrysler commented that a 10 percent AQL would force them to
incur additional testing costs. These costs would include a
one time investment of $1.70 million dollars for testing facilities
and equipment, and an additional $300,000 per year for employees.
Although comments were solicited on any increase in internal
audit testing which might be required by the 10 percent AQL, no
manufacturers responded with any information.
3. Analysis of the Comments
As can be seen from the preceding section, very little
37
-------
substantial comment was received from the manufacturers. Few
commentors replied in the detail requested in the NPRM, and only
minimal supporting cost breakdowns were given by those who provided
specific cost-related comments.
The EPA technical staff's analysis of these cost-related
issues will be presented on a subject by subject basis similar to
the preceeding section. In some cases the methodology used in the
recently finalized heavy-duty engine regulations will be utilized.
The costs which EPA will ultimately consider chargeable to
these regulations are only those which are necessary to meet the
requirements imposed by these rules and not necessarily the total
which the manufacturers stated they may spend.
A. Development and Emission Control Hardware
When considering the development and emission control hardware
costs related to the various aspects of this rulemaking package, it
is important first to consider the magnitude of the task. Based
upon the Technological Feasibility discussion in Issue G, costs
will be incurred in the following areas.
i. Redesign and Development Testing
Although the technology which will be employed to meet the
target emission levels is proven and well understood, development
costs will be incurred in improving the durability and reliability
of emission related components. This includes primarily the
catalyst, air injection, and EGR. Other related components which
may be affected to a lesser degree include emission-related carbu-
retor changes, electronic engine controls, and hardware used to
control evaporative emissions.
In addition, the EPA technical staff expects that some devel-
opment and testing will be done by the manufacturers to optimize
the engine/emission control system to achieve optimum engine
performance and fuel economy and emissions compliance for the
minimum cost.
On a per engine basis the costs of improving system reliabil-
ity and durability as well as engine optimization should not exceed
$5-$10 per engine.jV The EPA technical staff expects most of this
cost will be spent on improving the quality and durability of
materials used in the catalyst and hoses related to the EGR and air
injection.
In most cases optimization efforts by the manufacturer
would yield it a net benefit in terms of material costs saved or
marketability of the product. In any event, optimization costs
would be far less than $1.00 per engine, so this analysis will
assume the optimization cost is included in the $5-$10 estimate
cited earlier for redesign and development.
-------
ii. Emission Control Hardware
As stated in the technological feasibility discussion (see
Issue G), the EPA technical staff expects that most manufacturers
will use oxidation catalyst/air injection/EGR systems to comply
with the provisions of the proposed rulemaking. In addition, EPA
expects manufacturers to use electronic engine controls (EEC).
In most cases, changes to bring the vehicles into compliance
will be primarily the addition of electronic engine controls and
changes to catalyst volumes and loadings. In some cases air
injection will have to be added or the less effective pulse air
system will have to be replaced by a mechanical air pump.
To determine more specifically the costs related to bringing
the LDT fleet into compliance with the 1983 emission standards the
EPA technical staff studied the emission levels and corresponding
emission control hardware found in the*1980 LDT fleet.
As of February 1980, twelve light-duty truck manufacturers had
certified 47 engine families. Of these 47 families, 16 were
Federal families, 22 were Federal and California families and 9
were California only families.2/ Since engine families sold only
in California are certified to California emission standards, these
families will be eliminated from this analysis. When appropriate,
certification results for California engine families will be used
to indicate a possible emission control strategy which would bring
the Federal family into compliance.
The next twenty pages discuss for each manufacturer on an
engine family by engine family basis the steps which the manu-
facturer may need to achieve the approximate emission target levels
(.49 HC, 5.5 CO, 1.4 NOx) estimated by EPA.3/ At the end of the
discussion for each manufacturer is a table outlining certification
data for each engine family.JL3_/ Data used in this analysis is
actual certification data for 1980 LDTs. Costs in the Regulatory
Analysis will be based upon these strategies.
a. American Motors
BT9A1 - Engine family BT9A1 is a 151 CID engine employing EGR
and a pelleted catalyst. The HC level is below the target level
already, and the CO and NOx levels are within 0.1 g/mile of the
target levels. This engine family can be brought into compliance
with only minor engine calibration changes or electronic engine
controls.
Compliance Strategy: Use engine calibration changes or
electronic engine controls to decrease CO and NOx. Air injection
does not appear necessary.
CT3A1 - Engine family CT3A1 is a 258 CID/6 cylinder with air
injection/EGR/and a 160 cubic inch pelrteted oxidation catalyst.
-------
From the 4K emission levels CO appears to be the major problem.
Changing the catalyst volume and loading would increase CO oxida-
tion. In addition, a start catalyst may be a valid measure to
reduce CO emissions, however the full life durability of a start
catalyst is questionable.
Compliance Strategy: Increase the catalyst loading to 2.2 g
and increase the catalyst volume from 160 cubic inches to 200 cubic
inches. Add electronic engine controls to reduce cold start
emissions.
CT3H1 - Engine family CT3H1 is very similar to family CT3A1
and uses air injection, EGR, and a pelleted oxidation catalyst.
Some of the 4K emission levels are below the target levels.
Increases in the catalyst volume and loading, together with the
addition of electronic engine controls would decrease the emission
levels to near the target levels.
Compliance Strategy: Increase the catalyst loading to 2.2 g
and increase the catalyst volume to 200 cubic inches. Use electron-
ic engine controls (EEC) to reduce cold start and NOx emissions.
HT3A1 - Engine family HT3A1 is <± 304 cubic inch 8 cylinder
engine with air injection, EGR, and a pelleted oxidation catalyst.
None of the emission data vehicles met all of the target emission
levels. Increasing the catalyst volume and loading would likely
yield the desired emission reductions.
Compliance Strategy: Increase the catalyst loading to 2.515
grams and increase the catalyst volume to 260 cubic inches. Use
EEC to control cold start emissions and NOx.
NT3A1 - Engine family NT3A1 is a 360 cubic inch/8 cylinder
engine employing EGR, air injection, and a pelleted oxidation
catalyst. Although none of the emission data vehicles met the
target emission levels, compliance with the target levels appears
easily achievable through increasing the catalyst volume and
loading, and using the variable calibrations of EGR available with
EEC to control NOx emissions. These catalyst volume increases also
will be necessary to increase catalyst durability.
Compliance Strategy: Increase the catalyst volume to 260
cubic inches and increase the noble metal loading to 2.515 grams.
More NOx control can be gained through using modulated EGR or
engine calibrations. Use EEC to control cold start emissions.
I
b. Chrysler Corporation
r
,'
OTA-225-1-BCP - Engine family 225BCP is a 225 CID/6 cylinder
engine with air injection, EGR, and 2 small oxidation catalysts.
None of the emission data vehicles met any of the target emission
levels. This family will require increased catalyst loading and
volume as well as modulated EGR flow rates or a change in timing.
-------
Table F-l
AMERICAN MOTORS
Engine CID/
Family Cylinders
BT9A1 , 151/4
CT3A1 258/6
CT3H1 258/6
HT3A1 304/8
-C
NT3A1 360/8
Emission Control System Catalyst Data
AIR EGR CAT Volume - Loading
- X P 160 1.555g
X X P 160 2. -i- 1.555g
/j . i- tr
*fU
X X P 160 i.i 1.555g
^ . < YS"
X X P 160 zs>s 1.555g
d • ?
-------
Compliance Strategy: Increase the catalyst volume to 150 CID
and the catalyst loading should be increased to about 2.5 gram of
Pt per catalyst. Use EEC to reduce cold start emissions of HC and
CO, and permit the use of modulated EGR.
OTA-318-2-BCA - Engine family 318BCA is a 318 CID/8 cylinder
with air injection, EGR, and 2 small oxidation catalysts. A
compliance strategy similar to 225BCP should be adequate to lower
emissions to near the target levels. Some of the emission data
vehicles had 4,000 mile emission data values near or below the
emission target levels estimated by EPA. Modulated EGR will be
necessary to slightly lower the NOx emission level.
Compliance Strategy: Change the catalyst volume to 150 CID
and increase the platinum loading to 2.5 grams. Add EEC to reduce
cold-start emissions, and use modulated EGR to reduce NOx.
OTA-318-2-BEP - Engine family 318BEP is a 318 CID/8 cylinder
engine with air injection, EGR, and an oxidation catalyst. None of
the emission data vehicles met all of the target emission levels.
Increasing the catalyst loading and volume would be an efficient
means of reducing HC and CO emissions to near the target levels.
Electronic engine controls could be used to control cold start
emissions and would allow the use of modulated EGR to control
NOx.
Compliance Strategy: Increase the catalyst loading to approx-
imately 2.0 grams and increase the catalyst volume to 200 cubic
inches. Add EEC to reduce cold start emissions and NOx.
OTA-318/360-4BCP - Engine family 318/360 BCP represents
318/360 8 cylinder engines. As emission control systems these
engines employ air injection, EGR, and two small oxidation cata-
lysts. None of the emission data vehicles met all of the target
emission levels, although the HC and NOx values were fairly low.
To decrease the HC and CO emissions a larger more heavily loaded
catalyst should be used. Electronic engine controls could be used
to control cold start emissions and allow the use of modulated EGR
to control NOx.
Compliance Strategy: Increase the catalyst volume to 150
cubic inches and increase the platinum loading to 2.5 grams.
Use EEC to control cold start emissions and reduce NOx.
OTA-318/360-4BFP - Engine family 318/360 BFP represents a
318/360 8 cylinder engine. For emission control it uses air
injection, EGR, a start catalyst, and a regular underbody catalyst.
Only CO emissions are above the current target levels. Increas-
ing slightly the volume and noble metal loading in the underbody
catalyst will yield the required reductions and increased dura-
bility.
Compliance Strategy: Increase the catalyst loading to about
-------
Table F-2
CHRYSLER
Engine CID/ Emission Control System
Family Cylinders AIR EGR CAT
225 BCP 225/6 X X M
318 BCA 318/8 X X M
318 BEP 318/8 X X M
t
*
318/360 BCP 318 & 360/8 X X M
318/360 BFP 318 & 360/8 X X M
Catalyst Data 4K Emission Levels
Volume - Loading - Metals - Ratio HC
2 @ 45 -f.r.615g Pt 1 .97
A isfi 1.05
.55
.58
.61
2 @ 45 *s .615g Pt 1 .51
.76
.74
.60
.52
141 i* 1.44g Pt 1 1.0
A.n- 1.1
(.*•• -80
.80
.66
.66
2 @ 45 /r.615g Pt 1 .85
.78
.71
22/141 -.>J .9/2.88g Pt 1 .40
/ i/ .46
.35
CO
14
13
10
11
12
5.4
10.8
6.6
12.5
6.1
9.0
10.8
15.7
12.8
12.8
8.6
9.2
14.1
12.5
8.8
8.1
7.5
NOx
1.6
2.2
1.5
1.6
2.0
2.3
1.2
1.9
1.3
2.1
1.5
1.3
1.6
1.7
1.7
1.8
1.8
1.4
2.0
1.2
.90
1.1
-------
3.0 g and increase the volume to about 200 cubic inches. The
required reductions could be achieved by engine modifications but
this could be at the expense of fuel economy. Increasing the
volume and loading would also increase the emission control system
durability and diminish dependence on start catalysts. Use EEC to
reduce cold start emissions. Remove the start catalyst.
c. Ford Motor Company
4.9 NA - Engine family 4.9NA represents a 300 CID/6 cylinder
engine which uses air injection, EGR, and an oxidation catalyst.
One of the emission data vehicles met all of the target emission
levels with the current hardware, so it could be argued that no
incremental hardware cost is required. However, EPA expects that
Ford will minimize dependence on engine calibrations which may
be causing a fuel economy penalty. Therefore, Ford will most
likely use either EEC to reduce NOx and cold start HC and CO or
heavier noble metal loading to increase catalyst efficiency and
durability. We will conservatively assume that both are used.
Compliance Strategy: Increase noble metal loading to 2.5 g to
allow increased control and durability. Use EEC to reduce cold
start emissions.
5.0 NA - Engine family 5.0 NA represents a 302 CID/8 cylinder
engine which uses air injection, EGR, and an oxidation catalyst to
control emissions. Although none of the emission data vehicles met
all of the target emission levels the HC and CO levels are fairly
close already. Using a larger catalyst with heavier loadings would
yield the required emissions reductions together with no fuel
economy penalty related to HC and CO reductions. The use of
electronic controls will allow variable calibrations to reduce HC,
CO, and NOx emissions.
Compliance Strategy: Use the same catalyst as engine family
5.0 NB (150 CID/2.17 Pt and Pd in a 2:1 ratio). Use EEC to reduce
cold, start and NOx emissions.
5.0 NB - Engine family 5.0 NB is very similar to family 5.0
NA. It's emission data vehicles have HC and CO emission levels
below the targets and the NOx levels ^re easily achievable with
variable calibrations of the EGR available through EEC.
Compliance Strategy: Use EEC to reduce cold start and NOx
emissions.
5.8 M/6.6 NA - Engine family 5.8M/6.6 NA represents 8 cylinder
351 and 400 CID engines. These engines use air injection, EGR and
an oxidation catalyst. Although none of the emission levels from
the emission data vehicles met all of the target levels, several
met the HC and CO target levels. Using a slightly larger catalyst
with heavier noble metal loading should bring the desired HC and CO
reductions. NOx reductions should be achieved through modulated
EGR.
-------
Compliance Strategy: Use a 150 CID catalyst with 2.17 grams
of platinum and palladium in a 2:1 ratio. This is the same cata-
lyst as on family 5.8 WNG. Use EEC to reduce cold start emissions.
EEC will also allow the use of modulated EGR which will decrease
NOx.
5 .8 WNG - Family 5.8 WNG represents a 351 CID/8 cylinder
engine which use EGR, air injection, and an oxidation catalyst.
It's HC and CO levels are already below the target emission levels.
Further NOx control is available through EGR.
Compliance Strategy: To gain further NOx control use modu-
lated EGR.
d. General Motors Corporation
08F2A - Engine family 08F2A is a 250 CID/6 cylinder engine
which uses a pelleted oxidation catalyst, EGR, and a pulse air
system. Although none of the emission data vehicles met all of the
target emission levels, the HC and NOx levels are close to the
target levels. The CO levels are considerably above the emission
target level of 5.5g/mile. Three steps may be necessary to bring
this engine family into compliance: 1) replace the pulse air
system with a mechanical air pump 2) slightly increase the cata-
lyst loading and 3) add EEC to implement modulated EGR thus
reducing NOx. Adding more air will increase the HC and CO oxidation
and adding modulated EGR will assure adequate NOx reductions.
Compliance Strategy: Replace the pulse air system with a
mechanical air pump. Increase the catalyst loading to 2.515 grams
of Pt and Pd in a 5:2 ratio. Add EEC to reduce cold start and NOx
emissions.
08K4AA - Engine family 08K4AA represents 350 and 400 CID/8-
cylinder engines with EGR, air injection, and a pelleted oxidation
catalyst. Of the four emission data vehicles three met the target
HC and CO levels but none met the target NOx standards.
Compliance Strategy: Increase the catalyst loading to 2.515g,
and use EEC to implement modulated EGR and control spark timing.
Use EEC to reduce cold start emissions thus assuring HC and CO
compliance.
Q8K4G - Engine family 08K4G represents 350 and 400 CID/8-
cylinder engines which use only EGR and a pelleted oxidation
catalyst. None of the emission data vehicles met either the HC
or CO target levels and only one met the NOx target. The lack of
any air injection seems to be the cause of the shortfall in the HC
and CO areas. The 0.2 to 0.3 g/mile further NOx decrease required
can easily be achieved through modulated EGR.
/
Compliance Strategy: Add a mechanical air pump or at least
pulse air. Achieve required NOx reductions through modulated
-------
Table F-3
FORD
Engine CID/ Emission Control System
Family Cylinders AIR EGR CAT
4.9 NA 300/6 X X M
5.0 NA 302/8 X X M
5.0 NB 302/8 X X M
5.8 M/6.ii NA 351 & 400/8 X X M
5.8 WNG 351/8 X X M
Catalyst Data 4K Emission Levels
Volume - Loading - Metals - Ratio HC
150 2.17g Pt/Pd 2:1 .57
.87
.95
.72
.49
.44
128 1.85g Pt/Pd 2:1 .49
.59
.73
.61
.58
.51
150 2.17g Pt/Pd 2:1 .41
.59
128 1.85g Pt/Pd 2:1 .39
.56
.78
.39
150 2.17g Pt/Pd 2:1 .34
.33
.49
CO
5.0
8.8
10
8.1
4.2
5.2
5.0
3.5
6.9
7.6
9.6
4.1
2.3
4.8
2.8
6.1
12.0
4.5
2.9
3.9
3.7
NOx
1.1
1.9
1.4
1.3
1.6
1.3
1.6
1.8
1.6
2.0
1.3
1.5
1.5
1.6
1.9
1.5
1.5
2.0
1.6
2.0
2.0
-------
EGR. Add EEC to aid reductions of HC and CO during cold start.
08Y2A - Engine family 08Y2A is an 8 cylinder/305 CID engine
with a pelleted oxidation catalyst and an EGR system. None of the
emission data vehicles met the NOx target levels but two met the HC
level and one met the CO. From the emission data available it
appears likely that the required HC and CO reductions are attain-
able by the addition of some form of air injection and variable
timing. The required NOx reductions can be achieved through
modulated EGR.
Compliance Strategy: Add a mechanical air pump. Add EEC to
reduce cold start HC and CO emissions and allow implementation of
modulated EGR.
e. International Harvester Company
4-196 - Engine family 4-196 is a' 196 CID/4 cylinder engine
which use air injection, EGR, and a pelleted oxidation catalyst.
Certification data for this family shows the HC level is near
compliance, but the CO and NOx levels will require additional
reductions.
Compliance Strategy: Use EEC to reduce cold start HC and CO
emissions. Increase catalyst loading to 2.2 g. Gain NOx control
through modulated EGR.
V-304 - Engine family V-304 is a 304 CID/8 cylinder engine
using air injection, EGR, and a pelleted oxidation catalyst. The
emission data vehicles both had CO levels below the emission
targets. The HC levels are near the target levels and the NOx
levels are somewhat close. To reduce HC emissions a heavier
loading is necessary in the catalyst. To reduce NOx emissions
modulated EGR is necessary.
Compliance Strategy: Increase catalyst loading to 2.515 grams
of Pt and Pd in a 5:2 ratio. Add EEC to reduce cold start HC and
CO emissions. EEC will also allow the use of modulated EGR.
V-345 - Engine family V-345 is a 345 CID/8 cylinder engine
with air injection, EGR, and a pelleted oxidation catalyst. The CO
emission levels from both emission data vehicles achieved the
target levels and the HC level from one vehicle is below the
target level. This engine will require a heavier loading in the
oxidation catalyst to lower HC emissions and will require modulated
EGR to reduce NOx without a fuel economy penalty.
Compliance Strategy: Increase the catalyst loading from 1.555
to 2.515 grams of platinum and palladium in a 5:2 ratio. Add EEC to
control HC and CO cold start emissions and allow the use of modu-
lated EGR to reduce NOx.
-------
Engine
Family
08F2A
08K4AA
08K4G
V?
08Y2A
Table F-4
GM
CID/ Emission Control System Catalyst Data 4K Emission Levels
Cylinders AIR EGR CAT
250/6 PLS X P
-
350/8 X X P
400/8 X X P
350/8 - X P
400/8 X P
305/8 - X P
Volume - Loading - Metals - Ratio HC
260 1.555g Pt/Pd 5:2 .56
.43
.53
.70
260 1.555g Pt/Pd 5:2 .59
.41
.36
.41
260 2.515g Pt/Pd 5:2 .55
.43
.62
.57
.54
260 1.555g Pt/Pd 5:2" .42
.53
.49
.57
CO
15.0
12.0
9.7
16.0
8.0
4.5
4.1
5.2
12.8
7.5
11.3
10.2
13.0
6.2
10.2
5.5
8.6
NOx
1.2
1.1
1.8
1.9
1.9
2.0
2.0
1.5
1.7
1.6
1.6
1.5
1.3
1.6
1.8
2.0
1.8
-------
Table F-5
INTERNATIONAL HARVESTER COMPANY
Engine
Family
4-196
V304
V345
CID/
Cylinders
196/4
304/8
345/8
Emission Control
AIR
X
X
X
EGR
X
X
X
System
CAT
P
P
P
Catalyst Data
Volume
160
260
260
- Loading
1.555
1.555
1.555
- Metals
Pt/Pd
Pt/Pd
Pt/Pd
- Ratio
5:2
5:2
5:2
4K Emission Levels
HC
.5
.57
.61
.44
.57
CO
6.8
4.7
4.7
2.8
5.5
NOx
2.0
1.55
1.75
1.8
1.5
-------
f. Isuzu Motors
AlTB - Engine family AlTB represents a 111 CID/4 cylinder
engine with air injection and EGR. None of its current emissions
levels are below the target levels. If this engine were sold in
its California configuration nationwide all emissions would be in
compliance.
Compliance Strategy: Add an oxidation catalyst similar to
that used on the California version (160 CID, 1.555 g of Pt and Pd
in a 5:2 ratio). Add EEC to reduce cold start emissions of HC and
CO, and reduce NOx emissions through modulated EGR.
g. Mitsubishi Motors Corporation
G52T-F - Engine family G52T-F represents a 122 CID/4 cylinder
engine which uses EGR and a monolithic oxidation catalyst. None of
the emission data vehicles met all of the target emission levels
but one did meet the HC and CO targets. The California version of
this engine has emissions below the target levels for all three
pollutants. The major differences are the addition of a pulse air
system and engine calibrations.
Compliance Strategy: Add a pulse air system to reduce HC and
CO emissions. Add EEC to reduce cold start HC and CO emissions.
EEC will also allow the use of modulated EGR to reduce NOx.
G54T-F - Engine family G54T-F is a 156 CID/4 cylinder engine
which is sold only in the Federal version. It uses a pulse air
system, EGR, and a monolithic oxidation catalyst to control emis-
sions. Of the three emission data vehicles, one met all of the
target levels, two met the target HC levels, and two met the target
CO levels. Based on the data from the emission data vehicles, this
engine family can meet the new target levels with only engine or
emission control system calibration changes or the use of elec-
tronic engine controls. However, it is possible that a mechanical
air pump may be necessary to replace the pulse air system to assure
increased oxidation of the HC and CO which may be created by the
engine or emission control system calibration changes. This step
is only necessary as an added assurance that compliance will be
achieved.
Compliance Strategy: Use EEC to control cold start emissions
of HC and CO. Use of EEC will also allow NOx reduction . through
modulated EGR.
h. Nissan Motors
/
TL20F - Engine family TL20F represents a 119 CID/4 cylinder
engine which uses EGR, pulse air and a small monolithic oxidation
catalyst. None of the emission data vehicles met all of the target
emission levels. Only one met target NOx level and none met the HC
or CO targets. The California version of this family uses a
So
-------
Table F-6
ISUZU MOTORS
Engine CID/
Emission Control System
Catalyst Data
4K Emission Levels
Family
A1TB
A1TC
Cylinders
111/4
111/4
AIR
X
X
EGR CAT
X
X P
Volume - Loading - Metals - Ratio HC
1.3
1.5
1.4
1.4
160 1.555g Pt/Pd 5:2 .23
.26
CO
11.0
13.0
14.0
12.0
2.7
4.7
NOx
1.6
2.0
1.6
2.0
1.3
1.3
-------
Engine
Family
G52T-F
G54T-F
CID/
Cylinders
122/4
156/4
Table F-7
MITSUBISHI
Emission Control System Catalyst Data
AIR EGR CAT Volume - Loading - Metals - Ratio
- X M 61 2.5g Pd 1
PLS X M 61 2.5g Pd 1
4K
HC
.57
.75
.28
.16
.75
.28
Emission
CO
7.6
7.2
4.0
3.3
7.2
4.0
Levels
NOx
1.7
1.8
1.7
1.2
1.8
1.7
-------
mechanical air pump and a larger, more heavily loaded catalyst.
The California version of this family met all of the emission
targets by a comfortable margin.
Compliance Strategy: Sell the California version nationwide.
This would entail the replacement of the pulse air system with a
mechanical air pump, the use of an 80 cubic inch catalyst (as
opposed to 30) and an increase in the noble metal loading from .44
to 1.86 grams of platinum and palladium in a ratio of two to one.
Use the variable calibration capabilities of EEC to reduce HC, CO,
and NOx as needed.
i. Suzuki
LJ80 - Engine family LJ80 is a 49 cubic inch/4 cylinder engine
which is used in small land rover type vehicles. Its only emis-
sion control system is EGR. The EGR system allows this engine to
meet the target NOx level. Although the HC and CO levels are well
above the target levels, the addition of a pulse air system or, at
worst, a mechanical air pump should provide adequate air for
increased oxidation of HC and CO.
Compliance Strategy: Add a pulse air system or air pump to
reduce HC or CO emissions. If any additional reductions are
necessary these should be gained through the variable calibration
capabilities of EEC. The staff does not expect that an oxidation
catalyst will be necessary to pass the emission standards, but it
be necessary due to the idle standard.
j. Toyo Kogyo
OMAT - Engine family OMAT represents a 120 CID/4 cylinder
engine with a pulse air system, EGR, and an oxidation catalyst.
The emission data vehicle for this family met the HC target and
barely exceeded the NOx target. The CO level exceeded the target
by 0.7 g/mile. The California version of this family easily met
the target emission levels using only calibration differences from
the system being used Federally.
Compliance Strategy: Use the same control hardware as is
presently used, but use calibrations similar to those on the
California version or preferably the variable calibration cap-
abilities of EEC. Replace the pulse air system with a mechanical
air pump.
QWBT - Engine family OWBT represents a 140 CID/4 cylinder
engine which uses a pulse air system, EGR and a pelleted oxidation
catalyst. Of the three emission data vehicles used for Federal
certification, all met the HC target but only one met all three
emission targets. The one emission data vehicle which met all
three targets was sold in all 50 states. This engine family should
be able to meet all of the target levels with the hardware current-
ly in use.
-------
Table F-8
NISSAN MOTORS
Engine
Family
TL20F
CID/ Emission Control System
Cylinders AIR EGR CAT
119/4 PLS X M
Catalyst Data
Volume - Loading
30 .44g
- Metals - Ratio
Pt/Pd 2:1
4K Emission Levels
HC
.73
.83
.51
.62
CO
9.0
12.0
6.5
9.5
NOx
1.6
1.4
1.7
1.5
-------
Table F-9
SUZUKI
Engine
Family
CID/
Cylinders
Emission
AIR
Control
EGR
System
CAT
Catalyst
Volume - Loading -
Data
Metals - Ratio
4K
HC
Emission
CO
Levels
NOx
LJ80
49/4
1.0
12
1.4
-------
Compliance Strategy: Use the current hardware, but use EEC to
gain the reductions required to reduce emissions without a fuel
economy penalty. Replace the pulse air system with a mechanical
air pump.
k. Toyota
2F(F) - Engine family 2F(F) is a 258 CID/6 cylinder with air
injection, EGR, and a pelleted oxidation catalyst. Both of the
emission data vehicles met the HC target level but neither met the
CO or NOx target levels. The California version of this family has
emissions well below all target levels and uses a heavier loaded
catalyst of the same volume. The NOx control is gained through
minor calibration changes.
Compliance Strategy: Increase the catalyst loading to that
which is used in the California version. Use modulated EGR to
reduce NOx emissions.
20R(TC) - Engine family 20R(TC) is a 134 CID/4 cylinder engine
which was certified for sale in all fifty states in at least one
configuration. The emission data vehicle met all of the target
levels so this family can be considered in compliance.
Compliance Strategy: No action required to reduce emissions.
EEC could be added to reduce dependence on fuel consuming engine
calibrations which were used to reduce emissions for California
sales.
20R(TF) - Engine family 20R(TF) is the Federal only version of
the 20R(TC) family. This engine family uses air injection, EGR,
but no oxidation catalyst. None of the emission data vehicles met
any of the emission target levels. To bring this family into
compliance an oxidation catalyst will be necessary. This catalyst
should be similar if not identical to the catalyst used on family
20R(TC). In reality the likely outcome is the combination of
families 20R
-------
Table F-10
TOYO KOGYO (MAZDA)
Engine
Family
OMAT
OWBT
CID/
Cylinders
120/4
140/4
Emission Control
AIR
PLS
PLS
EGR
X
X
System
CAT
P
P
Catalyst Data
Volume
166
200
- Loading
1.82g
2.31g
- Metals
Pt/Pd
Pt/Pd
- Ratio
.84:. 16
.83:. 17
4K Emission Levels
HC
.40
.25
.37
.28
CO
6.2
2.8
7.3
6.5
NOx
1.5
1.1
1.6
1.6
-------
Engine
Family
2 F (F)
20R (TC)
20R (TF)
CID/
Cylinders
258/6
134/4
134/4
Table F-ll
TOYOTA
Emission Control System Catalyst Data
AIR EGR CAT Volume - Loading - Metals - Ratio
X X P 244 3.0g Pd/Pt 2:1
X X P 130 4.27g Pd/Pt 3:1
X X - - -
4K
HC
.44
.49
.16
.97
.65
.87
1.20
Emission
CO
8.7
9.5
5.0
13.0
10.0
12.0
14.0
Levels
NOx
1.8
1.7
1.2
1.7
1.6
1.8
2.0
-------
family 37PC were sold nationwide this engine type would come into
compliance.
Compliance Strategy: Sell family 37PC nationwide. A possible
option is the addition of EGR and a small air pump and catalyst,
but this methodology could be less fuel efficient.
Light-Duty Diesel Trucks (LDDT)
Currently three manufacturers GM, IH} and VW are certifying
light-duty diesel trucks for sale in the U.S. (see Table F-13).
Of the three families certified, two, those from IH and VW
already pass the target emission levels and will require no addi-
tional reductions.
The third, GM's 350, easily passe's the CO target level, but
does not pass the HC or NOx levels. This engine is basically the
same as that sold in GM's light-duty diesel passenger cars. The
passenger car version must meet 50,000 mile emission standards
which are more stringent than the LDDT target levels. The EPA
technical staff expects that the compliance strategy used to bring
the LDD passenger car into compliance will also be used on the
LDDT. This basically involves the addition of EGR and the redesign
of injectors and other minor engine modifications estimated to cost
$30 per engine.kj
iii. Other Control Strategies
Although the previous discussion has dealt almost exclusively
with oxidation catalyst/air injection systems there is another
control system available to the manufacturers. There is no doubt
that a three-way catalyst system could also be used to meet the
target emission levels. This three-way system would use a three-
way catalyst, electronic engine controls, and a feedback carbure-
tor. An approximate cost for this system on a 300-cubic inch
engine is shown in the attached table (see Table F-14).
iv. Summary
The strategies discussed for each light-duty truck family will
be used as a basic input for the economic impact chapter of the
regulatory analysis. The actual per vehicle cost for each of these
strategies will be estimated using the data and methodology found
in a report prepared under contract for EPA.JL2/ The estimates in
the report have been adjusted for inflation, increased real costs
of material, and a more realistic overhead and profit margin. The
profit and overhead margin used was derived in the 1984 heavy-duty
engine regulations summary and analysis of comments.
B. Certification
As a result of these new emission regulations, EPA expects
-------
Table F-12
VOLKSWAGEN
Engine
Family
37PF *
CID/
Cylinders
97/4
Emission Control System
AIR EGR CAT
X
Catalyst Data 4K Emission Levels
Volume - Loading - Metals - Ratio HC
- - 1.2
1.4
1.3
CO
6.1
6.5
5.7
NOx
1.8
1.8
1.5
Fuel 'Injection
-------
Table F-13
LIGHT-DUTY DIESEL TRUCK EMISSION DATA
Manufacturer
GM
IH
LDT
Family
09J9Z
SD-33T
Percent of
LDDT Sales
76.16
3.62
Crankcase
Control
Yes
No
Emission
Turbo
-
X
Control
EGR
-
X
System
FI
X
X
4K
HC
.84
.82
.45
.39
Emission
CO
2.0
2.0
2.2
1.6
Level
NOx
2.0
1.9
1.6
1.4
vw
DP
20.23
Yes
X
.32
.90
1.1
-------
Table F-14
Three-Way Control Strategy \J
Three-way Catalyst $181 2/
Feedback Carburetor Mods 7
Electronic Engine Controls 140
$328
less Oxidation Catalyst $117 _3_/
Air Pump 27
$184 4/
_!/ Incremental cost over a 1980 system.
2f A catalyst volume of 330 cubic inches with a noble metal
~ loading of 4.30g of Pt and .477g of Rh.
3/ A catalyst volume of 240 cubic inches with a noble metal
loading of 2.31 g Pt and 1.16 g Pd was used.
4/ 1980 dollars, includes profit at all levels.
-------
that most, if not all, LDT engine families will have to be re-
certified. This certification is expected to be a two step
process: preliminary deterioration factor assessment and testing
of emission data vehicles.
Preliminary deterioration factor assessment would most likely
entail the testing of 1-2 durability vehicles per family in an over
the road or test track usage. The cost of this testing will vary
by manufacturer depending on the type of vehicle used, the number
of durability vehicles per family and the useful life figure
determined.
In the economic impact analysis, EPA will use costs in 3 major
areas:
(1) Prototype vehicle,
(2) Mileage accumulation to 100,000 miles,
(3) Emission testing (28).
In most cases a 1975 EPA memo will be used to estimate these
costs.5J However the effects of inflation will be accounted for
by increasing the 1975 costs by 38.2 percent to yield 1980 dol-
lars.^/ Based on this memo and the inflation factor the costs of
preliminary deterioration factor should be approximately as shown
below:
1. Prototype vehicle: $ 35K
2. Mileage and maintenance
to 100,000 miles: 263K 3/
3. Testing (28) 12K 4_/
$310K per durability
vehicle
EPA expects that on the average each manufacturer will run 1-2
durability vehicles per family.^/ This durability vehicle program
will be conducted as part of the development program described
previously.
Emission data vehicles are much less expensive to develop and
run to 4,000 miles. Using the methodology and memo used above
these costs become:
1. Prototype vehicle $13,800
2. Mileage and maintenance
to 4,000 miles 8,800
3. Testing (2) 800
$23,400 per emission data
vehicle
EPA expects that on the average each manufacturer will run two
durability vehicles and four emission data vehicles per family.
-------
These costs are higher than those previously incurred during
certification primarily due to the unknown deterioration charac-
teristics of emission control systems past 50,000 miles.
C. Allowable Maintenance Provisions
Although no manufacturer commented on the costs of the new
allowable maintenance provisions, some costs will be incurred.
These costs are primarily related to research and development and
improved materials. Of all the items for which allowable mainte-
nance provisions were proposed, only two drew any real comment:
spark plugs and catalysts.
The analysis of the spark plug interval question (in Issue C -
Allowable Maintenance) indicates that the proposed interval is
within the range of current technology. Therefore, the staff
expects that the spark plug interval is obtainable at no extra
cost.
The catalytic converter interval of 100,000 miles drew the
most general comment and is where the greatest research effort will
be necessary. The $5-$10 which was estimated earlier as an R&D
cost should go toward assuring the durability of catalyst systems
for at least 100,000 miles. This $5-$10 does not include any
changes to catalyst loading which might be necessary to meet
the increased catalyst durability requirements. The costs for
increased loading were inherently addressed in the compliance
strategy discussion.
D. Useful Life Redefinition
1. Lower Target Levels
The change in the useful life definition will force the
manufacturers to achieve lower target levels in their compliance
efforts than they would have had the useful life remained at 50,000
miles/5 years. The table below compares the target levels for the
50,000 and 100,000 mile useful life periods. Both sets of targets
assume a 10 percent AQL.
Emission Target Levels 3/
50,000 mile useful life 100.000 mile useful life
~~HC0.57 g/mile 0.49 g/mile
CO 6.20 5.50
NOx 1.45 1.40
Although the NOx target level is not meaningfully affected,
the HC and CO target levels are substantially changed.
The cost of the lower target levels could be estimated by
determining the effect of the lower HC and CO targets on the
actual hardware used. Three main pieces of hardware are affected:
air pumps, catalytic converters, and electronic engine controls
(EEC).
-------
Of the 8 families which would have to add or upgrade their
present air injection systems, few, if any, would be able to
achieve even the 50,000 mile target levels without these systems.
This can be easily judged by comparing the HC and CO emission
levels of engines with and without air injection. Air injection
system changes and improvements will be caused primarily by the
more stringent emission standards and not the redefined useful
life.
Almost all families affected by these more stringent emission
standards will have to increase their catalyst volume and/or
loadings to meet the lower target levels. The actual amount of the
cost of the increased volume or loading directly attributable to
the increased useful life is difficult to estimate.
To estimate what amount of the catalyst related cost is
attributable to the longer useful life, it might be helpful to
study the HC and CO emission levels,and targets. Although HC
oxidation occurs directly in the catalyst, CO (one of the HC
combustion products) is the limiting pollutant, so CO will be
considered further. The required reductions in CO emissions can be
viewed in two segments: current levels to the 50,000 mile target
and further reductions from the 50,000 mile target to the 100,000
mile target. Using the concept of segmented reductions, the
catalytic converter cost related to the useful life can be esti-
mated. To estimate this cost the only additional piece of infor-
mation required is the current CO levels of the emission data
vehicles. Using the data gathered from EPA'" Certification Divi-
sion and averaged for each engine family, the average CO emission
level is currently about 7.8 g/mile (see Table F-15).
Total 100,000 mile CO reduction: 7.8 - 5.5 = 2.3
First 50,000 mile CO reduction: 7.8 - 6.2 = 1.6
Segmented Reduction: 1.6 = .70 :First 50,000 miles
2.3
2.3-1.6 = .30 :Second half of life
2.3
Using the methodology described above, the conclusion from
this analysis is that about 30 percent of the catalyst related
hardware costs could be attributable to the lower target standards
and thus the useful life. When computing the cost effectiveness of
the useful life, 30 percent of the catalyst cost computed in
Chapter V of the regulatory analysis will be included.
The third piece of hardware directly related to emissions
reductions are electronic engine controls (EEC). EEC are a rela-
tively new technology for maintaining or improving fuel economy and
engine performance while lowering emissions. EEC will contribute
substantially to lower cold start emissions due to variable timing
and lower NOx emissions in all driving modes as a result of
-------
Table F-15
Sales-Weighted Federal CO Levels
Manufacturer
AMC
Chrysler
Ford
^
7S
GM
Family
BT9A1
CT3A1
CT3H1
HT3A1
NT3A1
225 BCP
318 BCA
318 BEP
318/360 BCP
318/360 BFP
4.9 NA
5.0 NA
5.0 NB
5.8/6.6 NA
5.8 WNG
08F2A
08K4AA
08K4G
08Y2A
09J9Z
Sales
Weight
.0093
.0241
.0142
.0060
.0252
.0257
.0396
.0560
.0096
.0087
.0787
.1284
.0057
.0944
.0162
.0456
.1053
.1253
.0347
.0120
Avg 4K
CO Levels
5.6
9.7
5.2
7.1
11.4
12.0
8.3
11.6
11.9
8.1
6.9
6.1
3.6
6.4
3.5
13.2
5.5
11.0
7.6
2.2
Manufacturer
IH
Isuzu
Mitsubishi
Nissan
Suzuki
Toyo Kogyo
Toyota
VW
Family
4-196
V-304
V-345
SD33T
A1TB
A1TC
G52T-F
G54T-F
TL20F
LJ80
OMAT
OWBT
2F (F)
20R (TC)
20R (TF)
37 PF
DP
Sales
Weight
.0016
.0047
.0053
.0005
.0175
.0003
.0059
.0040
.0249
.0006
.0105
.0081
.0026
.0032
.0197
.0026
.0024
Avg 4K
CO Levels
6.8
4.7
4.2
1.5
12.5
3.7
6.3
4.8
9.3
12
6.2
5.5
9.1
5.0
12.3
6.1
1.0
Sales-Weighted Avg: 7.767 g/mile
-------
variable calibrations of EGR. These EEC will probably prevent the
need for the fuel consuming engine calibrations or start catalysts
used on many of the California engine families.
Although it cannot be stated unequivocally, it appears that
the manufacturers will choose the use of EEC to control cold start
emissions and maintain fuel economy. The start catalyst would be a
viable strategy for a 50,000 mile lifetime but its durability for a
full lifetime would be questionable.
The incremental cost of EEC for the second half of the life-
time could be estimated by determining the cost difference between
a start catalyst and a EEC system.
A start catalyst used by Ford on its California engine family
5.0 NG would cost the consumer about $47 Tj and an EEC system would
cost from $44 to $60 depending on production volume.8/ From this
analysis there appears to be no inherent additional cost increase
related to the EEC. A start catalyst has no significant cost
advantage and EEC will yield many other benefits to the manufac-
turer and owner for the full life. The EPA technical staff con-
cludes that no significant portion of the EEC is caused by the
lower target levels, so no EEC cost will be attributed to the
useful life redefinition.
2. Certification Costs
Since certification will be required for the full useful life
it is reasonable that manufacturers will run their durability
vehicles for the full useful life. If the full useful life is
100,000 miles, then one half of the mileage accumulation, mainte-
nance and testing costs for each engine family are attributable to
the full lifetime. Using the certification costs described pre-
viously, this cost comes to $137,500 per test vehicle or about
$275,000 per engine family.
3. Warranty
Although several manufacturers stated that per vehicle war-
ranty claims may exceedi $250, there is one unstated but basic point
in their analyses which leaves their comment open to some question.
In the past the manufacturers and vendors of emission related
components have designed and built these components knowing that
their responsibility ends at 50,000 miles. So these components
have been designed and built such that at 50,000 miles a small
percentage failure is acceptable. EPA expects that manufacturers
will not accept the potential warranty claims associated with
components designed for a 50,000 mile lifetime, but instead, will
redesign and eventually develop emission related components de-
signed for a full vehicle lifetime. Under this concept, the
manufacturers' warranty claims for a full lifetime should not
exceed that for the current 50,000 mile lifetime. This is true
because the acceptable failure rate at the average useful life
-------
should be the same as that for the 50,000 mile useful life.
As described briefly earlier, EPA expects that an average of
about $5-$10 per vehicle will be spent in this redesign and
development effort. This cost will primarily affect the catalyst,
air pump, EGR, and electronic engine control system, although some
lesser amounts may be spent on some other minor emission related
components. This $5—$10 estimate does not include the heavier
catalyst loadings described previously.
In any case, the manufacturers' estimates of increased war-
ranty claims should not occur because of the revision in the useful
life definition from that which was proposed. With the changes to
the proposed definition, manufacturers will no longer be liable for
engine rebuilds beyond their own warranty period.
E. Diesel Crankcase Control
Of the three light-duty diesel truck families currently
certified, the General Motors and Volkswagen families already have
closed crankcases. The remaining family, that from IH (actually
built by Mitsubishi), does not have a closed crankcase. This IH
family is turbocharged which may be the the reason the crankcase
has not been closed.
Mercedes Benz has closed the crankcase on both its naturally
aspirated and turbocharged light-duty diesel engines using a
cyclonic separator and a few hoses.2/ A local Mercedes dealer
estimated the replacement cost for these parts at $6 per engine,
which seems quite consistent when compared to the cost of other
crankcase control systems. A closed crankcase on gasoline-powered
light-duty vehicles costs about $2 per vehicle^/ and on heavy-duty
diesel engines this cost is approximately $10 per engine. 10/ The
estimate of $6 per vehicle seems reasonable in comparison to the
other control" systems, and will be used as the cost to close the
crankcase on the small number of IH diesels affected.
F. SEA Related Costs
The SEA related costs fall into three major areas: SEA
testing costs, self audit testing costs, and 10 percent AQL compli-
ance costs.
1. SEA Testing Costs
A decrease in formal SEA testing costs is expected as a result
of the change in the AQL and sampling plan. This will result in a
change in the average sample number per audit from 16 vehicles per
audit to 13 vehicles per audit. The change includes going from the
current batch sampling plan and a 40 percent non-compliance rate to
a sequential sampling plan and a 10 percent non-compliance rate.
On a per audit basis this amounts to about $1200. On a per manufac-
turer or per vehicle basis this savings is not substantial enough
-------
to receive any further consideration and will not be included in
the final cost analysis.
2. Self Audit Testing Costs
Although all major manufacturers were queried as to what
increases in self audit testing would be expected with a change in
the sampling plan and AQL, only Chrysler indicated that they would
require more testing. Chrysler estimated increased testing hard-
ware costs of $1.7 million and personnel costs of $300K per year.
Manufacturers' California audit testing data available to EPA
indicates that the manufacturers current compliance efforts at a 40
percent AQL are already yielding compliance levels near or surpass-
ing that necessary for a 10 percent AQL. This data is discussed in
the Technological Feasibility Issue (G). Based on this data the
EPA technical staff can see no need for any substantial increases
in self audit testing. However, because Chrysler responded in the
affirmative in the area of further testing, their costs will be
included in the final economic impact analysis.
3. 10 Percent AQL Compliance Costs
Due to the test to test variability of emissions data and the
more stringent 10 percent AQL, manufacturers will probably try to
achieve lower target emission levels than might be sought at a 40
percent AQL.
Using variability of 10 percent (HC), 20 percent (CO), and 25
percent (NOx) the EPA technical staff has estimated target levels
for a 40 percent AQL and a 10 percent AQL assuming in each case a
100,000 mile useful life.
Emission Targets _3/
40 Percent AQL 10 Percent AQL
HC 0.53 g/mile 0.49 g/mile
CO 6.4 5.5
NOx 1.7 1.4
To determine any hardware costs related to going from the 40
percent AQL to the 10 percent AQL it must be determined what if any
hardware changes may result from the slightly lower target levels
anticipated.
There are four basic pieces of hardware which must be con-
sidered: air injection, EGR, catalytic converters, and electronic
engine controls.
Of the families which need to add or upgrade an air injection
system, it does appear that perhaps one GM family (08Y2A) may be
able to meet the target HC and CO levels with the use of a pulse
-------
air system instead of a mechanical air pump if the 40 percent AQL
were retained. The actual cost differential between these two
systems is ($27-$4)=$23.9/ The 08Y2A family represents about 4
percent of all LDT sales so on a per vehicle basis the cost is
about §0.92 per engine.
Most of the NOx reductions required by the more stringent AQL
could be gained through changes in EGR or engine calibrations.
However, the EPA technical staff expects that the required reduc-
tions can be gained through the use of the EEC systems thus elimi-
nating the need for fuel consuming engine or EGR calibrations.
The diesel light-duty truck family produced by GM will require the
addition of EGR to meet the NOx targets. This EGR system would be
the same as that used on the light-duty diesel passenger cars and
cost between $8 and $15 per vehicle. The EPA technical staff will
be conservative and use the higher cost or about $15 per engine.
On a per LDT basis this is only $.60 per engine.ll/
The difference between the 40 and 10 percent AQL HC emission
targets is so small that it would probably not cause any substan-
tial change in catalytic converter volume or loading. The dif-
ference in the CO target levels is relatively larger, and it might
be argued that some of the catalyst volume or loading cost could be
saved with the 40 percent AQL. The EPA technical staff believes
that the manufacturers would probably use a slightly lighter
catalyst loading and not expend much effort in catalyst size
reduction. The EPA technical staff knows of no formula or rules of
thumb which relate noble metal loading and emissions reductions
explicitly. It seems reasonable that, at most, only 0.1 to 0.2
grams of noble metal could be saved in the catalyst. If one
assumes that this metal is Platinum , the cost of 0.1 grams would
be about $1.88. Thus lacking any other input this cost will be
used for the gasoline-powered light-duty trucks.
Electronic engine controls will probably be necessary to aid
in the reduction of cold start emissions. The EPA technical staff
has no doubt that EEC will be used regardless of the AQL because
there are so many benefits related to the use of EEC that the
manufacturers would probably implement these controls for the
marketing and fuel economy benefits alone. One other side benefit
of EEC is to allow the manufacturers to optimize catalyst volume
and loadings to achieve the greatest emission reductions at the
lowest cost.
4. Recommendations
The final cost figures used to compute the economic impact of
these regulations should be reevaluated based on the manufacturers'
comments.
-------
References
J7 This $5-$10 estimate was taken from the R&D costs for similar
component shown in EPA report 460/3-78-002, Cost Estimations
for Emission Control Related Components/Systems and Cost
Methodology Description, Leroy H. Lindgren, Rath and Strong,
Inc., March 1978.
_2/ Based on data gathered from EPA's Certification Data.
_3_/ See Chapter VII of the Regulatory Analysis which supports this
rulemaking.
_4/ Regulatory Analysis, Light-Duty Diesel Particulate Regula-
tions, OMSAPC, EPA, February 1980.
5J EPA memo, Light-Duty Vehicle Certification Cost, Daniel Hardin
to E.J. Brune, D.M. Kimball, and J.M. Marzen, March 1975.
6/ Based on CPI data received from the Bureau of Labor Statistic:
1976 - 4.8 percent, 1977 - 6.8 percent, 1978 - 9.0 percent,
1979 - 13.3 percent. Use of these percentages is very con-
servative.
7/ This catalyst is 52 cubic inches and is loaded with 1.2 g of
~ Pt and Pd in an 11:1 rat io.
8/ Sames reference as footnote one. The values on page 301 were
adjusted by an inflation rate of 26 percent over the three-
year period, the overhead/profit rate used was 29 percent and
costs were estimated at production volumes of 8 and 16 mil-
lion. Thus, ECUs are estimated to cost between $34-$47 and
sensors between $10 and $13 yielding a range of $44-$60 for
the simple system planned.
9/ Same reference as footnote one.
IO/ Based on Caterpillar Tractor Company's comment to the 1983 and
Later Model Year Heavy-Duty Engine Gaseous Emission Regula-
tions. |
ll/ Assumes this engine family, or one similar, comprises one-
third of the light-duty diesel trucks sold, or 4 percent of
all LDT sales.
12/ Cost Estimations for Emission Control Related Components/
Systems and Cost Methodology Description, Leroy H. Lindgren,
Rath & Strong, Inc, March 1978.
-71
-------
13/ Abbreviations used in this analysis: PLS - pulse air injec-
tion; M - monolithic oxidation catalyst; P - pelleted oxida-
tion catalyst.
7Z
-------
G. Issue: Technological Feasibility
1. Summary of the Issue
EPA's proposed HC and CO standards represent a reduction of 90
percent in the emission levels of uncontrolled baseline light-duty
trucks. This chapter explores whether these standards are tech-
nologically achievable within the context of revised certification
and auditing requirements and existing NOx and particulate control
requirements.
2. Summary of the Comments
Nearly all of the commenters agreed on the following point:
The proposed HC and CO emission levels are achievable provided that
the existing certification provisions and the current 40 percent
SEA Acceptable Quality Level (AQL) are left unchanged. Most
commenters argued, on the other hand, that in the presence of the
new useful life provisions and an AQL of 10 percent, meeting the
standards will be difficult or even impossible.
Several manufacturers provided estimates of how they expect
the revised useful life and/or 10 percent AQL requirements to
affect their ability to meet the standards. These estimates were
presented in two different ways. The first approach was to "ad-
just" the standards upward to a level which allegedly takes the
revised useful life and the 10 percent AQL into account. In the
second approach, the anticipated design targets were shifted
downward in such a way as to accommodate the new provisions. The
following paragraphs present the industry estimates.
General Motors (GM), International (IHC), and Volkswagen . (VW)
all offered increased standards which incorporate the effects of
the 10 percent AQL (IHC presented the numbers upon request by EPA,
but opposes the 10 percent AQL and any such adjustment of the
standard). The estimated levels follow:
HC g/mi CO g/mi
AMC (1.6 20.0
GM 1.4 20.0
IHC 1.1 17.1
VW 1.0 15.0
(Standards) (0.8) (10)
Toyota, GM, and IHC presented estimated design target levels
which would result from a 10 percent AQL. (GM's targets need to be
divided by the deterioration factors in order to be exactly compar-
able to the others.) Additionally, Toyota estimated the combined
effect of a full-life useful life (they used 100,000 miles) along
with a 10 percent AQL. In some cases, the targets corresponding to
the current NOx standard (2.3 g/mile) are included. All of the
estimates appear below:
-------
HC
(g/mi)
CO
(g/mi)
NOx
(g/mi)
Toyota (10% AQL, 50,000- .405 4.19 1.50
mile U.L.)
(40% AQL, 50,000- .462 4.86 1.65
mile U.L.)
(10% AQL, 100,000- .314 2.92 1.36
mile U.L.)
GM (10% AQL, 50,000- .482/DF 5.50/DF
mile U.L.)
(40% AQL, 50,000- .712/DF 8.84/DF
mile U.L.)
IHC (40% AQL, 50,000- .53 7.69
mile U.L.)
(40% AQL, 130,000- .34 5.49
mile U.L.)
(10% AQL, 130,000- .26 3.2
mile U.L.)
AMC (10% AQL, 50,000- .021 2.5 0.9
(mile U.L.)
(40% AQL, 50,000- 0.42 5.0 1.8
(mile U.L.)
Chrysler expected an 18.5 percent increase in the stringency
of the standards to accompany a 10 percent AQL, although no quan-
titative support was provided. Similarly, AMC stated that the
target emission levels under a 10 percent AQL constraint would
approach those of 1983 light-duty vehicles, which, they add, is
"clearly not the intent of Congress."
The interaction of the existing LDT NOx and particulate
standards with the standards proposed in this rulemaking was the
subject of comment, particularly from current and prospective
diesel manufacturers. During the comment period for this proposed
rule, a 0.2 g/mile particulate standard had been proposed for
light-duty trucks in a separate rulemaking. Since then, a final
standard of 0.26 g/mile has been promulgated. Thus, it was with
respect to the lower, proposed standard that the comments were
made. The current NOx standard is 2.3 g/mi.
AMC claimed that if a 10 percent AQL is instituted, the 2.3
g/mi NOx standard would probably need to be relaxed if they are to
avoid a shift to three-way catalyst systems. With respect /to
diesel-powered light-duty trucks, Cummins stressed the inter-
dependency between NOx and particulate control (e.g., increased EGR
for NOx reduction can increase particulate emissions). And
several manufacturers producing or considering producing light-duty
diesel trucks expressed strong concern that the proposed particu-
late standard will be impossible to achieve with known technology.
-------
Chrysler, GM, and AMC all claimed that a 0.2 g/mi particulate
requirement would probably present an insurmountable obstacle.
Some of the commenters offered their assessment of what
approach they would pursue in trying to meet the proposed stan-
dards. General Motors presented the most complete scenario. The
following hardware changes and improvements are described as GM's
"best effort," based on their California experience. (The proposed
standards are equivalent, they say, to California's 1980 50,000-
mile standards.)
GM expects to have to add air injection to their 5.0-, 5.7-,
and 6.6-liter engines; the pulse-air system on the 4.1-liter engine
will be replaced with an air pump. Regarding the catalyst, a
heavier noble metal loading, special converter shell, and struc-
tural improvements will be required at a minimum for the sake of
increased durability. Finally, GM anticipates that engine modifi-
cations to reduce oil consumption will be necessary. Three-way
closed-loop systems are not being contemplated.
Chrysler, however, does expect that they would use three-way
systems to achieve the design target emission levels. However,
they state that in the absence of a full-life useful life they
would use a small 22in^ start catalyst with a 40 g/ft^ platinum
loading in addition to their main catalyst. AMC as well suspects
that three-way catalyst systems will be required for them to meet
the current NOx standard if the proposed HC and CO standards plus a.
10 percent AQL are implemented. Volkswagen also believes that a 10
percent AQL would require them to use three-way systems. In the
presence of a 40 percent AQL, however, they would expect to use
only an oxidation catalyst and EGR. VW would like to avoid the use
of an air pump because intermittent NOx reduction, they say, can
occur in a catalyst during portions of the driving cycle. If
additional air injection is required to meet the HC and CO stan-
dards, this NOx reduction will be lost and the current NOx standard
will be difficult to achieve.
IHC states that emission levels as low as their estimated
design targets have been seen, but they doubt that they could
consistently reach such levels in certification. This company
claims their ability to comply with the regulations is dependent to
a large degree on the performance of outside vendors, which supply
much of IHC's hardware. By having to depend on other companies for
technological improvements in their hardware and systems—and
perhaps for some engines as well—IHC feels they have a reduced
ability to define their technological feasibility.
3. Analysis of the Comments
The general approach of this section will be twofold. First,
we will compare the 1980 California certification emission levels
of current light-duty trucks with EPA estimated target levels. Our
targets are based on those offered by the industry in conjunction
-------
with our own analysis. Second, we will investigate the achievabil-
ity of these target levels by 1980 engine families and what
control strategies we would expect to be necessary. By following
this approach we will be able to treat the comments in a general
way rather than engage in a comment-by-comment rebuttal.
The detailed EPA staff analysis which resulted in the pro-
jected emission target levels is presented in Chapter VII of the
Regulatory Analysis document titled, "Cost Effectiveness." In-
dustry data about emission variability were combined with antici-
pated deterioration factors to arrive at projected target levels
which reflect full-life useful life requirements in the context of
a 10 percent AQL. The reader is encouraged to consult this anal-
ysis, which in effect constitutes much of our response to the
comments which concerned emission variability and effects of a
full-life useful life and an AQL of 10 percent on low-mileage
emission target levels.
a. Gasoline-Powered Light-Duty Trucks
Our analysis of current emission performance will center
around engine families certified for California (or 50-state)
sales. Since California's 1980 LOT emission standards are con-
siderably tighter than the 1980 Federal standards and are in the
same range as the 1983 standards proposed here,* one would expect
that California-certified engines would offer some insight into the
emission control strategies required to meet EPA's proposed stan-
dards.
There is one reason for caution when one makes comparisons
to California vehicles. Manufacturers have warned that in com-
plying with California standards they have seen a loss in the fuel
economy of their vehicles. However, as we discuss in Issue L of
this document, "Fuel Economy," the California control approaches
amounted in most cases to "quick fix" modifications of Federal
vehicles—modifications that have stressed easy compliance over
fuel-economy considerations. Thus, from control technology stan-
dard standpoint, an analysis using 1980 California trucks is
compromised by the fact that manufacturers will most likely begin
to emphasize control systems which minimize fuel economy for future
systems.
Despite this drawback we believe that a look at the California
control systems and emission levels will be helpful. Emission
rates for California vehicles indicate what is readily attainable
using current catalyst systems. By placing side-by-side the
California numbers and the anticipated Federal targets, we will be
able to clarify at least generally where the feasibility problems
lie and how compelling they seem to be.
* 1980 California LDT standards range from .41-.9 g/mi for HC,
9-17 g/mi for CO, and 1.0-2.3 g/mi for NOx, depending on vehicle
weight and (for NOx) type of certification. Federal 1980 standards
for LDTs are 1.7, 18, and 2.3 g/mi for HC, CO, and NOx.
-------
The 4,000-mile emission levels for the 1980 LOT engine fam-
ilies that were certified either for California only or for all 50
states are presented in Table G-l. These values are "undeterior-
ated," that is, they have not been multiplied by deterioration
factors, the process which would yield the certification emission
levels. As they are, the numbers represent the actual low-mileage
emission test results for the certification fleet (California
plus 50-State).
But the question arises as to whether this sample, consisting
solely of California engines, is representative of the entire LDT
fleet. To a large degree it is representative, because only a few
1980 federal-only trucks have engines which were not sold in
California trucks. These are not exceptional polluters, so conclu-
sions about emissions feasibility drawn from the California data
can be reasonably extrapolated to the LDT fleet as a whole.
The numbers which we will compare with those of Table G-l are
the staff's projected 1983 emission target levels, assuming that
the proposed regulations are adopted:
HC: .49 g/mi
CO: 5.5 g/mi
NOx: 1.4 g/mi
It is worthwhile to stress again that these target levels (from
Chapter VII of the Regulatory Analysis) are computed on the basis
of actual LDT emission variability and under the assumptions of a
full-life useful life and a 10 percent AQL.
As one studies the data of Table G-l, a demarcation quickly
appears between the emission levels of the smaller, generally
4-cylinder imported vehicles and the levels of the domestic
trucks. The foreign vehicles show considerably better emission
performance; in fact, none of the 29 test vehicles exceeded the
estimated HC target, only two exceeded the CO target and 5 met or
exceeded the NOx target (just one of these actually exceeded it).
Because the domestic and imported trucks demonstrate such marked
differences in emission performance, we will treat them separately
in our discussion! It seems reasonable to concentrate on the
American manufacturers as presenting the most difficult feasiblity
problems.
It is clear from Table G-l that GM, Chrysler, IHC, AMC, and
Ford will have to do development work above and beyond what was
necessary for 1980 California certification in order to reach EPA's
estimated targets. Effort will need to be focused primarily on
improvements in CO and NOx. Hydrocarbon control is much less of a
problem, and control of CO will generally improve HC simultaneous-
ly. Nine of the 44 domestic test vehicles (20 percent) are above
the HC target, but only 3 exceed the target by more than 0.1 g/mi.
Carbon monoxide levels as recorded in Table G-l are higher
than the targets in 19 American test vehicles, and most of those
-------
Table G-l
1980 California Certification Results
Manufacturer
AMC
Chrysler
Ford
Engine Family
(Displacement, CID)
HT-3V1
(304)
NT-3A1
(360)
CT-4W1
(258)
BT-6C1
(151)
OTA-225-1-BXP
(225)
OTA-318/360-4BCP
(318/360)
OTA-318/360-4BFP
(318/360)
4.9 NA \J
(300)
4.9 ND
(300)
5.0 NB
(302)
4K-Mile Emissions (g/mi)
5.0 NG
(302)
HC
.38
.28
.59
.57
.43
.40
.28
.18
.18
.30
.45
.22
.22
.44
.33
.40
.37
.27
.49
.44
.23
.17
.24
.36
.41
.59
.42
.70
.40
.31
.35
.35
CO
2.9
2.8
9.9
9.1
7.3
4.4
3.8
2.5
1.9
3.7
5.2
1.8
7.0
8.3
5.6
7.0
8.8
5.8
4,2
5.2
2.1
5.7
6.2
2.4
2.3
4.8
4.2
9.5
6.7
1.6
2.1
2.0
NOx
0.92
1.5
2.0
1.8
1.0
1.4
1.5
1.3
1.2
1.7
1.4
1.3
1.2
1.4
1.8
1.9
1.7
1.7
1.6
1.3
1.9
1.7
1.6
1.5
1.5 11
1.6 I/
1.4
1.5
1.1
i i
1 1>4
( 1.4
1.4
-------
Table G-l (Cont'd)
1980 California Certification Results
Manufacturer
GM
IHC
Isuzu
Nissan
Mitsubishi
Toyo Kogyo
Engine Family
(Displacement, CID)
5.8M/6.6NA
(351/400)
5.8 WNG
(351)
08F2A
(250)
08K4AA
(350)
V-304
(304)
V-345
(345)
4-196
(196)
A1TC
(111)
TL20C
(119)
GT5-C
(121.8)
,
OMAT
(120)
4K-Mile
HC
.39
.56
.20
.34
.33
.39
.43
.30
.41
.64
.77
.55
.26
.23
.23
.23
.22
.26
.24
.18
.14
.26
.19
.26
.20
.24
Emissions
CO
2.8
6.1
2.4
4.5
13.8
9.2
8.7
7.3
5.2
5.0
6.7
8.9
4.7
2.0
2.7
2.7
4.3
3.0
2.2
2.6
2.6
3.8
2.5
3.5
3.2
-2.4
(g/mi)
NOx
1.9 I/
1.5 I/
1.3
2.0 if
.98
1.5
1.4
1.0
1.5
1.3
2.0
1.7
1.3 I/
1.0
1.3
1.3 I/
.98
1.0
1.0
1.1
1.4
1.2
1.1
1.1
1.2
1.1
-------
Table G-l (Cont'd)
1980 California Certification Results
Engine Family
Manufacturer (Displacement, CID)
OWBT
(140)
Toyota 2F(C)
(257)
20R(TC)
(133.6)
VW 37PC
(97)
37PF
(97)
11
4K-Mile
HC
.24
.23
.25
.18
.20
.15
.12
.20
.16
.16
.16
.16
.19
.28
Emissions
CO
2
2
2
3
3
1
2
2
1
4
2
2
3
7
.5
.6
.8
.3
.0
.9
.3
.2
.9
.9
.0
.0
.8
.6
(g/mi)
NOx
1
1
1
1
1
1
1
1
1
1
1
1
.2
.2
.1
.2
.5
.4
.90
.0
.1
.2
.4
.4
.70
.1
11
I/
12 .28 7.6 1.1
I/ 50-state certification.
-------
miss the target significantly (their average emission level is some
40 percent over the target). The 19 vehicles represent eleven
engine families.
Completing the picture of current emission control perfor-
mance, we find that NOx seems to present the most difficult obsta-
cle to compliance with the proposed regulations. Fully 32 of the
44 American vehicles in Table G-l, representing all but two Cali-
fornia and 50-state engine families, meet or exceed the NOx target
(24 vehicles actually exceed it). The average emission level is 12
percent above the target; 5 vehicles are 29 percent or more above.
It is plain that the level of effort which the American
manufacturers applied to meeting the 1980 California standards will
not be sufficient to comply with these 1983 regulations. We cannot
agree, however, with the comments which argue that the task EPA
places before the manufacturers with this rulemaking will only be
accomplished with new technology, high cost, and major compromises
in fuel economy, if it can be accomplished at all. On the con-
trary, we find abundant reason to believe that—even in the context
of a full-life useful life and a 10 percent AQL—the proposed
standards are achievable by all LDT manufacturers, for the most
part with current control systems. The next paragraphs support
this conclusion.
In order to reduce the HC and CO emissions of their federal
vehicles to California levels the manufacturers have for the most
part relied on improved catalyst efficiency and/or additional
auxiliary air injection. Some have also introduced small "start"
catalysts or calibration changes to improve cold-start emission
performance. The cold start controls present special problems
which we will cover shortly. In general, though, we believe that
manufacturers will again pursue, to a large extent, catalyst
improvements and increased air for their 1983 trucks, just as they
did for California trucks in 1980. Light-duty truck catalysts are
an especially fertile area for emission improvements, as the next
paragraphs demonstrate.
Increasing the internal volume and/or the noble metal loading
of catalysts can serve two crucial functions. The first of these
is the improvement of catalyst efficiency, as defined as the
percent reduction in mass exhaust emissions available through use
of the catalyst. Increased catalyst volume implies that a greater
surface area of substrate material is exposed to the exhaust flow,
improving the opportunities for pollutant molecules to react.
Independent of but related to converter sizing is the quantity of
noble metals which are applied to the substrate surface. An atom
or a group of atoms forms an active catalyst site, and the amount
of reaction that can take place is dependent on the number of these
sites that are exposed to the exhaust flow. Although there is
obviously a limit, a general rule is that the greater the catalyst
loading, the greater can be the HC and CO emission reductions.
-------
In addition to improving the conversion efficiency, increased
converter volume and noble metal loading make the catalyst and its
operation more durable. This is an obvious advantage in light of
the proposed extension of light-duty truck useful life. The
durability of the converter structure—i.e., the physical integrity
of the substrate—is improved because the greater mass and volume
make it less likely that critical temperatures will ever be
reached (1700°F for gamma alumina substrates). The progressive
loss of surface area associated with occasional high-temperature
excursions, and hence the loss of active sites and efficiency, is
thus minimized. An additional advantage is gained by increased
noble metal loading in that more active sites are initially avail-
able. Thus, even if a number of sites are rendered inactive
through the life of the vehicle by lead or phosphorous poisoning,
loss of substrate surface area, or agglomeration (catalyst atoms
migrating toward each other, effectively reducing the number of
sites), a significant number will remain available. In these ways
converter sizing and loading can act to extend the functional
lifetime of the catalyst. While both of these approaches are
clearly available to the manufacturers, we expect increased loading
to be the prevalent choice since manufacturers will probably wish
to limit the number of catalyst canister sizes that they make.
Putting aside for the moment the improvements in catalyst
durability, we believe that the additional HC and CO control
available through more heavily loaded (or larger) catalysts will
easily be sufficient to meet the proposed HC and CO standards.
As is often the case, EPA is not able to quantitatively support
this statement because actual data from LOT's equipped with
prototype 1983 catalysts are obviously not yet available. However,
one finds that current catalysts—even on California trucks—are
for the most part much more lightly loaded than what we would
expect to be required to comply with these regulations. Economic
pressures have clearly encouraged the use of the least expensive
systems necessary to meet the California standards. Coupling this
observation with the Agency's accumulated knowledge of catalyst
technology, we conclude that a considerable margin for improvement
in catalyst efficiency exists in most cases.
In some instances, such catalyst improvements will require
additional air injection. Most current families of light-duty
trucks are equipped with air pumps, which add air to the exhaust
stream in order to assure the presence of sufficient oxygen to
maximize oxidation in the catalyst. A few truck families use a
modulated "pulse-air" system, and a few more have no auxiliary
air. We expect that most trucks will have to incorporate addi-
tional air injection in order to make full use of the improved
catalysts, particularly American-made vehicles with larger engines;
if there is presently none, an air pump may need to be added. The
implications which increased use of air injection hold for LDT fuel
economy are addressed under Issue L, "Fuel Economy."
GM will perhaps be most affected by a need for additional air
injection. We agree that their 4.1-liter engines, which currently
-------
use the "pulse-air" system, will need air pumps. While a large-
selling 50-state family using a 5.7-liter engine (08K4AA) already
uses air injection in California applications, GM is probably
correct in predicting that auxiliary air will be necessary for the
remainder of the 5. 7L engines as well as for the 5.0L engine
family.
While we have been able to project that catalyst improvements
and increased air injection can themselves bring many current
trucks into compliance with the 1983 HC and CO requirements, the
actual picture will be somewhat different. This is because we are
anticipating a major shift by 1983 to the use of electronic engine
controls (EECs) as a part of LDT emission control packages. The
introduction of EECs, already popular in passenger cars, should
occur at least partially as a result of inherent conflicts between
cold-start emission control and fuel economy. Both conventional
ways of reducing HC and CO prior to catalyst light-off suffer from
drawbacks. Small, rapid light-off catalysts are subject to quick
deterioration due to their proximity to the engine. The alterna-
tive approach, adjusting the timing to speed up the heating of the
catalyst, causes a fuel economy penalty all during the vehicle's
operation (these calibration changes are the cause of much of the
loss in fuel economy experienced in the 1980 California fleet
compared to the Federal fleet). EECs offer a way of varying the
timing, returning the engine to a more fuel-efficient calibration
after the catalyst begins to work. Thus, while working to reduce
th heavily-weighted cold-start portion of a vehicle's emissions,
EECs can simultaneously avoid a loss in fuel economy (Issue L,
"Fuel Economy" explores this issue further).
Although meeting the existing NOx standard of 2.3 g/mi will
become more difficult with the implementation of the proposed rule,
compliance is within the reach of all domestic (and foreign)
manufacturers using current technology NOx control. That current
technology is exhaust gas recirculation (EGR), a system found today
on nearly all light-duty trucks. An increased flow of recycled gas
will probably be required for most U.S.-made trucks, though this
increase should not go far beyond what has been required for
California compliance. Table G-l demonstrates that while Califor-
nia systems often fall short of the EPA target, few of the levels
are grossly above that target. Our conclusion, is that minor
increases in EGR rates or none at all will be sufficient for
most engine families to remain within the existing 2.3 g/mi stan-
dard. A small number of engine families, concentrated primarily in
the larger CID range, may require a significant amount of EGR.
This expectation is tempered by the fact that a shift away from the
heavier engines will probably occur in the next few years, a topic
that is taken up in detail in Chapter V of the Regulatory Analysis.
As a final note, EGR has negative implications for fuel economy,
but modulation with an EEC helps to minimize this. We address this
issue in section L of this document, "Fuel Economy" and conclude
that on a fleet-wide basis, the net fuel economy loss due to NOx
control will be minimal.
-------
Manufacturers may opt in some of these latter cases to apply
three-way catalyst/feedback carburetor technology, but this should
be a rare occurence; we doubt strongly that these regulations will
require a move to three—way technology for any manufacturer. This
doubt is based on two facts. First, NOx control efforts on
California trucks have been aimed at meeting relatively high
standards (in most cases, 2.0 or 2.3 g/mi, depending on whether
deterioration is calculated on the basis of 50,000 or 100,000
miles). This makes it unlikely that many of these engines have
reached the limits of NOx reduction through EGR. Second, it is
generally in the small number of engine families which exceed 340
CID in size that the more difficult NOx problems seem to be con-
centrated. Again we point to the anticipated downward shift in
engine size, which will act to reduce the number of engine families
with higher NOx emissions. Our conclusion is that the degree of
NOx control necessary to meet these regulations, barring possible
exceptional cases, is available to all manufacturers through
increased EGR.
We have primarily directed the foregoing discussions of HC,
CO, and NOx feasibility at American-manufactured engines. The
basic strategies for meeting the proposed standards, i.e., catalyst
improvements, more air injection, additional EGR, and EECs, apply
as well to the manufacturers of imported LDTs. However, the
already lower emissions of the imports demonstrate that the neces-
sary degree of improvement in the existing control systems is
small; in many instances no improvement over California technology
appears to be necessary at all to achieve EPA's estimated targets.
Slight improvements in catalyst conversion efficiency (as discussed
earlier) may be desired for lower CO for some families, but this
approach is probably not necessary. Thus, the need for air injec-
tion (or additional air injection) will not be widespread. (This
would mean that VW will probably not have to sacrifice the NOx
control they get in the absence of an air pump). Regarding NOx,
again very little is required, but a slight increase in EGR above
that seen in California certification may be necessary in some
cases. In any event, compliance with the regulations by all of
the foreign manufacturers seems assured with California-type
control systems, perhaps upgraded with electronic controls.
There was a great deal of comment that indicated that the
effect of a 10 percent AQL and a full-life useful life would be
greater than what we have concluded in -this analysis. As a check
on the reasonableness of our feasibility analysis, we have per-
formed a separate analysis which uses actual 1980 California LDT
production audit data. (This work has been placed in the public
docket and is titled "Analysis of California 2% Audit Data.")
We have applied the estimated full-life deterioration factors* to
the audit emission data and calculated the rate at which these
vehicles fail to meet the proposed Federal standards. The analysis
used emission data from 7 tests on IH vehicles, 81 on Chryslers,
and 262 on Fords, and 263 on GMs, totaling 613 tests. The re-
1.4 for HC, 1.3 for CO, and 1.04 for NOx.
-------
suiting failure of the proposed standards occurred at a rate of 5.4
percent for HC, 7.8 percent for CO, and 8.8 percent for NOx.
Further, if one exceptionally failure-prone Ford engine family is
removed from the calculations, the failure rates drop to 5.0, 7.1,
and 4.4 percent for HC, CO, and NOx. We can conclude that with
absolutely no effort toward meeting the proposed regulations, a
significant number of American made California LDTs already do and
they would quite easily pass a 10 percent AQL audit. It would seem
that our original analysis based on the target levels results in a
conservative view of LDT feasibility. This look at actual produc-
tion performance is strongly supportive of our conclusions that
compliance with these regulations is achievable.
A final issue relating to feasibility deserves discussion.
While we factored into our analysis the effects of a full-life
useful life from the standpoint of initial zero-mile emission
requirements, we have not yet discussed the feasibility of de-
signing emission control systems which actually function for the
120,000-odd-mile lifetime of an LDT. Since durability is not a
major problem for air pumps and EGR systems, the discussion should
focus on the longevity of catalysts in LDT applications.
In developing a separate argument, AMC presented data for 4
catalyst-equipped light-duty vehicles. The data consist of peri-
odic emission results extending out to 100,000 miles. (The data
has been plotted in Figures G-l through G-4). While the data
points are somewhat scattered, there is in three vehicles a clear
pattern of linear deterioration of HC, CO, and NOx emission deteri-
oration (or improvement in the case of NOx). This is what we would
expect to occur in a normally functioning catalyst; i.e., a gradual
loss of efficiency with no evidence of severe damage. (One vehicle
seems to have suffered engine problems late in its life — HC and
CO emissions skyrocketed yet NOx went down.) On those three
vehicles, the catalyst was still functioning well—that is, there
had been no abrupt failure—even after 100,000 miles of use. It is
even more telling that these long catalyst lives are seen on
1975-76 vehicles equipped with early technology automotive catalyst
systems.
Further evidence of catalyst durability comes from a recent
EPA program in which 8 catalyst-equipped cars with an average
accumulated mileage of 104,480 miles were emission tested (as
reported in "Evaluation of Restorative Maintenance and Catalyst
Replacement on Exhaust Emissions from Eight Very High Mileage
Passenger Cars in St. Louis," available from the public docket
associated with this rulemaking). The vehicles were tested in 5
configurations: in an as-received condition, after correction of
maladjustments, after complete tune-up, with the catalyst removed,
and with a. new catalyst. An upward jump in HC and CO emissions
when the catalysts were removed demonstrates that the catalysts
were clearly still functioning even at these high mileages.
Further, after maladjustments and disablements were corrected and a
tune-up performed, half of the vehicles were able to meet the
-------
FIGURE G-l
(a)
(b)
(c)
2.0
l.S
|
s »••
a
0.5
041-200
' *:• •'*'
* * * * * * * *
o. ZOOM. 40000. eeoeo. ooooo. 10*
i«.
12
) CGM/HID_
o S
0
4.
2.
* •',
041-200 * 4
4 * *
4 * * 4
* 4 ***
4
). 20000. 40oae. cooeo. eoooa. lee
4,e
COM/MI J
N W
b B
i
i.»
« •.«!
041-200
**»»*.** »**
*
« *
1. 20000. 4CO^. £0000. 663BO. 10(»
MILEAGE MILEAGE MILEAGE
FIGURE G-2
i r
• .0
r-t
•^
O
0.4
•.2
(a)
D40-ICIO.
*
*****
4 • -t.» * *
4> * * * " '
10.
B.
cfi
^
8*
2.
•. 2CCC9. 4ccco. caoae. .eocao. iccoa "-J
(b)
« D40-19KL
4> *
» 4.
» *
4
* * « «
* *
2ccao. 4ocoe. (ooao. eoaoo. loot
4O
3.0
~
12(>
w
1
1.0
10 •
(c)
040-1SKL
t
* * 4
»
* * * * 4 '*
ZBoeo. 4oaoa. «eoo9. eeooo. leeoa
MILEAGE MILEAGE MILEAGE
-------
(a)
FIGURE G-3
(b)
(c)
_ 9.Cr
z:
o
a. 2
D4B-22R ^ *
«
*
» *
* , * *
»
*
). 2x&£3. 40ua. £oeeb. ecooa. 100
in
8
£6
2.
99 C
048-22R
* * *
« « «
* *
* *
3.6
»-<
i.e
D49-22R »
*
« *****
* . * * * * * * * *
20980. 40008. £0009. 99009. 100OB ""j. 20009. 40003. £0009. 60O30. 10OO
MILEAGE MILEAGE MILEAGE
OQ
o
(a)
FIGURE G-4
(b)
(c)
1 *
e e
a-
0 4
e 2
*
D56-1 OKA *
t *
» * * * » *
«. 2Kiea 4aeo0. £»0o0. eoooo. leo
29.
19.
£
2 10.
0
o
5.
69 0
*
D5B-10KA
*
+ * * *
* *
* * * * * *
+
20009. 46009. £0000. 60009. |OOC
4 «>
3 e
I..
0
i.e
a.e
10 i
D5S-10KA
»»»»»:****** »
*
20009. 40COO. £OOM. OOOOO IWX
MILEAGE MILEAGE • MILEAGE
-------
federal emission standards with their original catalyst — roughly
the'fraction that would be expected to have met the standards as
they came off the production line under the production practices
existing at the time (designing for the average vehicle to meet the
standard).
Thus, evidence clearly exists that passenger car catalysts,
while designed to operate only 50,000 miles, commonly last twice
that long. The light-duty truck catalysts that we expect to be
used to comply with these regulations will be, as discussed ear-
lier, inherently more durable due to sizing and loading improve-
ments. It is a reasonable conclusion, then, that long-lived
catalyst technology will be available for 1983 light-duty trucks.
b. Diesel-Powered Light-Duty Trucks
The issue of feasibility is very different for diesels. They
generally show low HC and CO emissions; but because of the higher
combustion temperatures (as compared to gasoline engines) NOx
values tend to be higher. In addition, one must be concerned with
particulate emissions, standards for which have recently been
published (Particulate Regulation for Diesel-Fueled Light-Duty
Vehicles and Light-Duty Trucks, 45 FR 14496, March 5, 1980).
As a basis for comparison, we will use the same estimated
low-mileage targets that we used for gasoline LDT's. This is
because emission variability and deterioration information is
very sparse for diesel LDTs. Since the HC and CO deterioration
factors will on the average be smaller for diesels, the gasoline
truck targets will be conservative and will provide an additional
"margin of safety" in the following feasibility analysis. NOx
deterioration, on the other hand, may be more rapid in diesels, and
it might appear that the diesel NOx target should be lower to
compensate for this. But tfre 1.4 target is already a conservative
estimate because of the very high NOx variability that was used.
We feel justified in using the gasoline target for diesels as
well. For convenience we repeat the targets (g/mi): HC - 0.49, CO
- 5.5, NOx - 1.4.
We have collected the 1980 certification data for light-duty
diesel trucks and for GM's diesel-equipped Oldsmobile passenger car
in Table G-2. The passenger car is included since it uses an
upgraded version of the same basic engine that the GM trucks do and
was tested at relatively high inertia weights (4,000, 4,500, and
4,750 Ibs). Its emission performance should reasonably represent
that of a LOT equipped with that engine configuration. ! '
Carbon monoxide values are all well below the target. HC
falls below the HC target in all but one case. The exception, the
GM diesel truck, uses an engine configuration different than that
used in the Oldsmobile,* probably the 1979 configuration. The
* GM used different injectors and introduced EGR on the Olds-
mobile.
-------
Table G-2
1980 Light-Duty Diesel Certification Emission Data
(Undeteriorated Gaseous Emissions, g/mile)
Manufacturer
GM LDT
GM LDV
(Oldsmobile)
IHC LDT
VW LDT
Engine Family
(Displacement, CID)
09J9Z
(350)
03J9ZG
(350)
SD-33T
(198)
DP
(90)
4K-Mile
HC
0.84
0.61
0.82
0.28
0.30
0.34
0.17
,,0.45
0.39
0.32
Emission
CO
2.0
1.9
2.0
1.2
1.1
1.1
1.2
2.2
1.6
0.90
(g/mi)
NOx
2.0
2.0
1.9
1.8
1.6
1.4
1.7
1.6
1.4
1.1
-------
passenger car has had to meet a tight 0.41 g/mi standard in
1980 for the first time, and its performance clearly indicates that
the truck's HC could greatly be reduced by using the new engine
configuration. It appears, then, that there is nothing inherent in
current diesel LDT engines that will prevent their meeting EPA's
HC and CO targets.
With respect to NOx on the other hand, additional development
work will be required of the manufacturers in some cases in order
to achieve the estimated 1.4 g/mi target. This does not seem to be
a large problem. One finds in Table G-2 that, if we exclude the GM
truck, all current certification vehicles lie in the 1.2-1.8 range
(in fact, the VW engine falls well below the targets on all three
pollutants). We again make an exception for the GM LDT, assuming
that the cleaner Oldsmobile LDV engine will be used in the truck
applications. Focusing, then, on the IH truck and the Oldsmobile,
an increased rate of EGR should provide the degree of NOx control
necessary for the vehicles to reach the vicinity of 1.4 g/mi. This
is a reasonable conclusion on the basis of EPA's observation of
diesel passenger cars, which indicates relatively low rates of EGR
are currently used and hence, a range of control remains available
to manufacturers.
The final item pertaining to diesel light-duty trucks is the
question of particulates. The adoption of full-life useful life
and a 10 percent AQL affects the difficulty of meeting particu-
late standards as well as HC, CO, and NOx. However, the staff
finds that this increased difficulty is overcome by the relatively
less stringent NOx standard applicable to light-duty trucks in
comparison to light-duty vehicles.
The particulate standard of 0.6 g/mile which will be applic-
able for 1983 and 1984 was derived so as to be achievable in the
context of a 1.0 to 1.5 g/mi light-duty vehicle NOx standard (the
actual amount depending on the outcome of NOx waiver requests). By
comparison, the light-duty truck requirement of 2.3 g/mile is quite
generous. Where EGR is being used to control NOx, there is a
direct relationship between NOx and particulates such that as NOx
goes up, particulates go down. The following analysis shows that
the less stringent 2.3 g/mi NOx standard allows sufficient flexi-
bility to meet the 0.6 g/mi particulate standard for light-duty
trucks at a 10 percent AQL and full-life useful life.
The Regulatory Analysis for the light-duty diesel particulate
regulations included some data which had been supplied by General
Motors in their May 1979 application for waiver from the 1981-1984
NOx emission standards for light-duty diesel engines.JY This data
illustrates the relationship between changing particulate and NOx
emissions for 3 GM test vehicles, all in the 4,500 Ib. inertia
weight class. This data will be used to characterize the relative
amount of particulate/NOx tradeoff which could occur, all other
things being the same.
-------
In analyzing the feasibility of meeting the 0.6 g/tni par-
ticulate standard, the staff found that GM would have the most
difficulty with the standard ._2/ Therefore, although the amount of
tradeoff may vary between manufacturers, we believe that if the
available flexibility is sufficient for GM it will also be suf-
ficient for other manufacturers.
The tradeoff flexibilities in going from a NOx level of 2.3
g/mi (the light-duty truck standard) to 1.5 g/mi (the maximum
light-duty vehicle standard) amounts to 36 percent for car "A," 54
percent for car "B," and 52 percent for car "C." That is, all
other things being equal, going from 2.3 g/mile NOx to 1.5 g/mile
NOx causes the particulate emissions of these three engines to
increase by the given percentages. Since, for the case at hand,
all other things are not equal, some or all of this available
"cushion" will get used up. Our desire is to determine if indeed
this available "cushion" is sufficient for light-duty trucks to
meet the same standard as light-duty'vehicles.
There are three areas where "cushion" is needed. First is the
increase in emissions associated with increased road load horse-
power and inertia weight of light-duty trucks. The staff analysis
for the light-duty particulate package concluded that these in-
creases amount to approximately 20 percent.3/ Second, a "cushion"
to cover the added full life useful life requirements is needed.
Using a maximum deterioration rate of i.l for 50,000 milesj4/, this
adds another 10 percent (to go from 50,000 miles to 100,000 miles).
Lastly is the amount of "cushion" needed to go from a 40 percent
AQL to a 10 percent AQL. This can be estimated using the method-
ology of Chapter VII of our Regulatory Analysis for this rule-
making, once an estimate of production variability is known. The
staff analysis for the light-duty particulate package determined
that particulate variability should be quite lov.5j Using the
relatively low variability characteristic of hydrocarbons as an
estimate, the data in Chapter VII of the regulatory analysis
indicates an 8 percent "cushion" associated with changing the AQL.
The total "cushion" needed is then 20 + 10 + 8 = 38 percent.
The required "cushion" compares very favorably to that
available from the three test vehicles. Two of those vehicles
(cars "B" and "C") exceed this requirement by a wide margin. The
third, car "A", also has an adequate "cushion." In calculating the
36 percent available, no allowance was made for the fact that the
actual NOx levels would be somewhat below, rather than at the
standard. It is in the nature of the NOx/particulate relationship
that at smaller NOx levels, the amount of tradeoff increases.
Also, from the point of view of the needed amount of 'cushion', the
38 percent is an estimate of what should be the maximum need. For
example, the 20 percent change associated with inertia and and road
load differences between light-duty vehicles and light-duty trucks
was a conservative estimate from data showing an average change of
16-18 percent. Therefore, the staff concludes that the feasibility
analysis associated with the light-duty particulate regulations
-------
remains valid, even in the context of a 10 percent AQL and a full
life useful life.
In 1985, we expect that new NOx control requirements will
become effective for LDT's and heavy-duty engines. Although
it is not necessary to discuss the effect of such a standard
on LDT diesels (that is a topic for the NOx rulemaking itself),
it is germane to look at how the current 2.3 g/mi NOx standard
will affect light-duty diesel trucks if it is retained after
1985. Again, the light-duty diesel particulate final rulemaking
considered NOx in the establishing of the second-stage (1985)
particulate standards, 0.20 g/mi for LDVs and 0.26 g/mi for
LDTs. The 30 percent higher standard for trucks is the result
of an analysis which assumed that a stricter NOx standard* would
indeed be implemented in 1985, removing the "safety margin"
which the trucks would have previously enjoyed (as per the pre-
ceding paragraphs). The adjustment in the particulate standard was
made to address just such a situation. In the absence of a 1985
NOx reduction, there should remain something of the 1983-84
margin. Thus, for the post-1985 time frame, we see no incon-
sistency between a 2.3 g/mi NOx standard and a 0.26 g/mi partic-
ulate standard.
c. General Responses
IHC made a comment which relates to feasibility yet is not
technical in nature—that vendor dependency constrains their
ability to change technology. We recognize the special problems
associated with relying on outside suppliers. However, in this
particular case, we do not agree that the resulting problems need
be severe. For example, the required technological improvements
are relatively minor and should not in themselves cause undue
delays in vendor deliveries. Also, since this is an industry-wide
action, the vendors, too, should be well aware of the changes at a
very early stage.
A general overview of the comments and our response in the
area of feasibility reveals significant disagreement between EPA
and the commenters regarding the effect that a 10 percent AQL will
have on emission targets. No commenter performed an analysis
approaching EPA's in thoroughness. We are confident that our
conclusions are based on sounder grounds than those of the com-
menters. For example, AMC was noted earlier as claiming that the
target emission levels under a 10 percent AQL would approach those
of light-duty vehicles. However, this position was based on
nothing more than an "assumption" that the 10 percent AQL would
lead to target levels that were one-half of those for the current
* They assumed that an LDT NOx standard comparable in stringency
to the 1.0 g/mi LDV standard (1.0 g/mi presumes no further NOx
waivers beyond 1985) would be enacted. This is a "worst case"
scenario, since the 1985 NOx standard is not likely to be so
stringent.
-------
40 percent AQL. .Our analysis shows that this is not true. The HC
and CO targets for a 10 percent AQL are 92 percent and 86 percent
of those for a 40 percent AQL, respectively. Targets used by
individual manufacturers may vary somewhat from those we have
estimated, but not by the amount claimed by AMC.
It should also be recognized that the delaying of in-use
durability requirements has removed one of the complicating
factors and has thus improved the situation of the manufacturers.
4. Recommendation
The staff concludes that no technological barriers exist to
prevent compliance with these regulations by all light-duty truck
manufacturers. We recommend that the proposed standards be re-
tained .
-------
References
\J Regulatory Analysis of the Light-Duty Diesel Particulate
Regulations for 1982 and Later Model Year Light-Duty Diesel
Vehicles. Figure IV-1.
_2/ Ibid, pg. 42.
_3/ Ibid, pg. 57.
4/ Ibid, pg. 36.
5/ Ibid, pg. 34.
-------
H. Issue: Selective Enforcement Auditing
In the July 12, 1979 NPRM, EPA proposed to revise the current
Selective Enforcement Auditing (SEA) program for light-duty trucks
(LDTs). The proposed SEA replaces the 40 percent Acceptable
Quality Level (AQL) currently in effect with a 10 percent AQL and
substitutes a sequential sampling plan for the batch sampling plan
that is presently in Subpart G. There were also several other
proposed changes to the current LDT SEA program.
The major portion of the LDT manufacturers' comments on the
proposed LDT SEA concerned the 10 percent AQL; therefore, a large
part of this summary and analysis is dedicated to that issue.
Comments on other aspects of the SEA proposal are addressed fol-
lowing the discussion of the 10 percent AQL.
Acceptable Quality Level
1. Summary of the Issue
In brief, this issue can be stated as follows: What Accept-
able Quality Level (AQL) should be promulgated in the final rule?
The AQL represents the percentage of light-duty trucks (LDT) within
a given population which will be allowed to exceed the emission
requirements. The Clean Air Act does not specify the precise AQL
to be applied to an assembly-line testing program like SEA.
A 10 percent AQL reflects EPA's view that the statute requires
every vehicle to be warranted to meet the emission standards
while allowing a 10 percent failure rate to account for measurement
error and inevitable quality aberrations.
EPA promulgated an initial 40 percent AQL for its current SEA
program because at the time the SEA regulation went into effect
(1976), the industry was building vehicles and trucks to meet
previously established standards on the average. A 40 percent AQL
assures that, for a vehicle population assumed to have a skewed-
normal distribution, vehicles within this population will comply
with standards on the average. In order to have brought the
light-duty engine families into compliance with a 10 percent AQL,
manufacturers at that time would have had to add additional emis-
sion control equipment to retain their certificates of conformity.
EPA's intent in promulgating an initial 40 percent AQL for its
current SEA program was to provide light-duty vehicle and truck
manufacturers the time and flexibility to bring all their motor
vehicles into conformance with the standards on a reasonable
schedule. This schedule is to parallel efforts to improve fuel
economy.
In the LDT Notice of Proposed Rulemaking (NPRM), the Agency
proposed a 10 percent AQL as part of the total compliance strategy
outlined in the proposal. EPA indicated that the 10 percent AQL
-------
could be met with costs not unreasonably burdensome to the manufac-
turers. Comments on cost associated with meeting a 10 percent AQL
were requested in the proposal.
2. Summary of the Comments
All the manufacturers and organizations that responded
to the NPRM opposed the implementation of a 10 percent AQL for
SEA. Most of the comments concerned the legal, technological and
economic reasons why a 10 percent AQL should not be promulgated.
Very little data were provided relating to the actual technological
and economic considerations associated with meeting a 10 percent
AQL.
All commenters made one or more of the following major points:
the 10 percent AQL is inconsistent with the intent of Congress and
the Clean Air Act; the 10 percent AQL would have the effect of
lowering the standards to a point where the emission reductions are
greater than the reduction mandated by Congress; the 10 percent AQL
is inconsistent with the 40 percent AQL used in the light-duty
vehicle (LDV) Selective Enforcement Auditing program; and the 10
percent AQL is not feasible due to testing and production vari-
ability. In addition, commenters gave various other reasons why a
10 percent AQL should not be put into effect: it is inconsistent
with certification requirements which imply "averaging"; it will
cause penalties in fuel economy; the air quality benefit of a 10
percent AQL has not. been calculated; and no cost estimates have
been made concerning the increased costs, in the areas of produc-
tion testing, emissions hardware, and fuel consumption, involved
with implementing a 10 percent AQL.
Most manufacturers stated that due to the above reasons the
AQL should be revised to 40 percent in the final rule. Some gave
examples of a 40 percent AQL sampling plan that EPA could adopt for
the final rule.
a. Light-Duty Truck Manufacturers
General Motors (GM)
GM stated that it is opposed to a 10 percent AQL sampling plan
for SEA. GM believes that Congress intended, in the 1977 Clean Air
Act Amendments, that production vehicles meet the emission stan-
dards on the average.
In its SEA discussion, GM directed most of its arguments
towards supporting the concept of averaging for production line
testing. These arguments included statutory language; the Draft
Regulatory Analysis discussion which GM claims to be based on
averaging; ambient air quality considerations; various past Con-
gressional Committee reports and statements of EPA Administrators;
consistency with certification requirements; and the analysis of
the baseline testing program.
-------
GM also stated the emission design targets necessitated by the
10 percent AQL would be below those justified in the environmental
and economic impact analysis. GM provided an analysis based on its
1979 production varability. GM estimated the design targets as
.482/DF g/mile for HC and 5.50/DF g/mile for CO. GM also submitted
an alternative sampling plan which determines compliance based on
the mean emissions of a population. GM advocates the use of this
sampling plan because it incorporates the averaging concept which
GM believes is consistent with the intent of Congress.
Ford Motor Company (Ford)
Ford stated that it opposed a 10 percent AQL sampling plan for
SEA. Ford believes that, as a matter of law and good engineering
practice, a 10 percent AQL is inappropriate for production-line
auditing.
Ford argued that it is not technically feasible to meet
a 10 percent AQL and an averaging concept is necessary for produc-
tion line testing. Ford submitted an analysis of the design target
as a function of the ratio of the standard deviation to the Low
Mileage Target. The Low Mileage Target equals the emission stan-
dard divided by the deterioration factor. On the basis of this
analysis, Ford claims that it would not be technologically feasible
for it to comply with the 10 percent AQL.
Ford also stated that the 10 percent AQL is inconsistent with
the certification process and the intent of Congress. Ford
developed an AQL sampling plan similar to the one proposed by
EPA. Ford's plan incorporated a 40 percent AQL. It suggested that
this plan be adopted in the final rule because a 40 percent AQL is
consistent with the requirements of the Clean Air Act and the
current SEA program.
Chrylser Corporation (Chrysler)
Chrysler commented that the 10 percent AQL represents a
"considerable increase in stringency" over the 40 percent AQL and
would result in a greater than 90 percent reduction from baseline.
Chrysler estimated that the 10 percent AQL would cause it to lower
its design targets by approximately 18 percent. The estimate is
based on an analysis in which Chrysler assumed a lognormal dis-
tribution and stated the coefficient of variation is .2. This
manufacturer stated that the Clean Air Act does not require every
vehicle to meet the emission standards throughout its useful life.
Rather, Chrysler believed the Act "compels" averaging because of
the Section 202(a)(3)(A)(ii) statement requiring reductions "- .
.from the average of the actually measured emissions. . ." It
stated that a 40 percent AQL was instituted for the current SEA
program "to avoid an unreasonable economic impact on the in-
dustry." Chrysler stated that the several SEA program were
evidence that the industry is not advanced enough in its production
-------
practices Co contend with a 10 percent AQL and that the 10 percent
AQL was therefore unwarranted and unsubstantiated. Chrysler also
stated that in the Regulatory Analysis EPA recognized the similar-
ity between light-duty vehicles and light-duty trucks; therefore,
"It would make little sense to attempt to extract different quality
levels from essentially identical types of vehicles."
Chrysler advocated adopting a 40 percent AQL because of
economic and technological considerations and because it would
satisfy the Congressional intent of averaging.
American Motors Corporation (AMC)
AMC stated that it opposed a 10 percent AQL sampling plan for
SEA because it "would increase the stringency of the proposed
numerical standard greatly in excess of the 90 percent reduction
mandated by Congress and is inconsistent with the Congressional
intent of averaging".
AMC believes that the functional reliability of exhaust
emission components necessary to meet a 10 percent AQL is not
"practical and affordable" in volume production of vehicles. AMC
stated that the design targets .necessary for it to reach the 1983
LDT standards with a 10 percent AQL sampling plan are .21 g/mile
for HC and 2.5 g/mile for CO. AMC estimated that the design
targets necessary to comply with a 10 percent AQL were 50 percent
lower than the design targets necessary to meet a 40 percent AQL.
However, an explaination of how this figure was derived was not
provided with the analysis. AMC stated that a 10 percnet AQL would
have the effect of increasing the stringency of the NOx standard to
a point that would require more sophisticated emissions control
technology (e.g. three-way catalyst). AMC estimated that it would
cost it and its vendors several million dollars to assure a 10
percent AQL.
International Harvester Company (IHC)
IHC opposes the 10 percent AQL sampling plan for SEA because
it is claimed to be beyond the statutory authority provided to EPA
and because it would allegedly cause an adverse economic burden on
the manufacturers. IHC believed that Congress did not intend a 10
percent AQL be promulgated because that AQL imposes emission
standards more stringent than those required by certification. IHC
stated that the legislative history of the -Clean Air Act indicates
that vehicles need only meet standards on the average. In iaddi-
tion, IHC felt that the imposition of a 10 percent AQL would have
an adverse economic impact on the light-duty truck industry, so it
should be relaxed as it was in the light-duty vehicle SEA regula-
tions. IHC provided an analysis in which the design targets
necessary to meet a 10 percent AQL for SEA were estimated to be .26
g/mile for HC and 3.2 g/mile for CO.
-------
IHC recommended that EPA retain the present 40 percent AQL to
avoid imposing a much more stringent emission standard than Con-
gress intended.
Cummins Engine Company, Inc. (Cummins)
Cummins believes that the potential impact of * 10 percent AQL
must be carefully studied. Cummins stated that due to moderate
test-to-test variability and vehicle design tolerances, manufac-
turers would need to design their products to meet an emission
level which is significantly lower than the legislated requirement.
Volkswagen (VW)
VW stated that to comply with a 10 percent AQL for the pro-
posed emission standards it would be necessary to use the closed
loop controlled three-way catalyst concept. VW estimates that a 10
percent AQL would require the average emission level to be about 20
to 30 percent below the engineering goals currently used by VW.
This reduction is solely because of the measuring uncertainties.
VW did not provide an analysis to support its estimate of the
percent reduction necessary to meet a 10 percent AQL.
VW recommended that "until government together with industry
can minimize these uncertainties, the AQL should remain at 40
percent."
Toyota Motor Co. Ltd. (Toyota)
Toyota does not agree with EPA1s proposing a 10 percent AQL
instead of a 40 percent AQL because this manufacturer believes that
a 10 percent AQL is not consistent with the intent of the Clean Air
Act.
b. Other Commenters
Motor Vehicle Manufacturers Association (MVMA)
MVMA's main comment was that the 10 percent AQL is not con-
sistent with an averaging concept for determining compliance with
standards. Its arguments were based on the legislative history of
the Clean Air Act, statements of past EPA Administrators, the
averaging concept embodied in certification regulations, ambient
air quality studies based on averages, the statutory language in
Sections 202(a) and 202(b), and the inclusion of averaging concepts
in the Regulatory Analysis for the NPRM.
U.S. Department of Commerce (DOC)
The DOC commented that there is no rationale in the NPRM for a
10 percent AQL. DOC stated if a 10 percent AQL is established
engines and emission control systems must be designed to emit
-------
levels of pollutants considerably below the levels permitted by
certification. DOC believes that a 10 percent AQL will increase
component costs significantly, may require more sophisticated and
expensive emission control hardware and would involve a fuel
economy penalty. DOC did not supply an analysis supporting these
statements. DOC recommended the retention of the 40 percent AQL
for LDT.
U.S. Council on Wage and Price Stability (COWPS)
COWPS stated that the compliance cost estimates contained in
the EPA Regulatory Analysis make no allowance for any increased
manufacturing costs associated with the 10 percent AQL.
COWPS also suggested that the 10 percent AQL may go beyond the
statutory mandate because it will require either a very low
variance in the test results for production vehicles or an average
level of emissions which is low relative to the standard. For
these reasons COWPS recommended a cost-benefit analysis be per-
formed before a 10 percent AQL is established.
3. Analysis of the Comments
Since many of the manufacturers and organizations responding
made similar comments on the AQL issue, each of the major comments
will be discussed under a separate heading in this section for
purposes of clarity. For further information relating to the 10
percent AQL issue, reference is also made to the economic impact
and the cost-effectiveness studies in Chapter V and Chapter VII of
the Regulatory Analysis and to the discussion of the technological
feasibility of the emission standards in the Summary and Analysis
o f Comment s.
a. The 10 Percent AQL is Consistent with Congressional
Intent
When reviewing the comments to the NPKM on SEA for light-duty
vehicles in 1976, the EPA Office of General Counsel (OGC) reached a
finding that "...Congress intended that, eventually, every car
coming off the assembly line should meet the emission standards
established under Section 202." A copy of the memorandum contain-
ing this finding (General Counsel Opinion No. 76-4) is available in
the Public Docket for this Rulemaking. OGC acknowledged that a
phasing in of this requirement was appropriate to avoid imple-
menting SEA in an unreasonably burdensome manner, so long a!s the
ultimate goal of full compliance is not abandoned. As explained [in
the LDV SEA preamble (41 FR 31474, July 28, 1976), auto manufac-
turers argued that implementation of a 10 percent AQL would have a
disastrous economic impact on the industry, since it would result
in a loss of certification for a majority of engine families. A 40
percent AQL was therefore established to implement SEA in a manner
not unreasonably burdensome to the affected manufacturers. This
initial approach was designed to "provide manufacturers the time
-------
and flexibility to bring all their vehicles into conformance with
the standards on a reasonable schedule" (41 FR 31475).
Section 206(b) of the Clean Air Act provides for SEA testing
of light-duty trucks. EPA maintains the position that there is a
specific legal basis for requiring every LDT coming off the
assembly line to meet emission standards. The full text of the EPA
General Counsel memorandum, mentioned above, explains that the
language of the Clean Air Act and the relevant legislative history
support an "every car" approach to compliance with emission
standards.
The ultimate goal of every vehicle and engine complying
with emission standards is also supported by the U.S. General
Accounting Office(GAO). The GAO did not take issue with EPA's
legal interpretation of the Clean Air Act on this matter and
recommended that the current LDV SEA program be revised to "...re-
quire a Federal emission standard compliance rate more indicative
of the current rate for car configurations tested, which is well in
excess of the 60 percent passing rate required." (GAO Report CED
78-180, p. 28).
b. The Relationship Between the Standards and a 10 Percent
AQL
Section 202(a)(3)(A)(ii) of the CAA states, in pertinent
part, "... regulations . . . applicable to emissions from vehi-
cles or engines manufactured during and after model year 1983, in
the case of HC and CO, shall contain standards which require a
reduction of at least 90 percent . . . from the average of the
actually measured emissions...during the baseline model year."
Pursuant to this requirement, EPA conducted a test program on 1969
model year light-duty trucks (the last model year before imposition
of HC/CO standards for light-duty trucks). Using the sales-
weighted average emission levels obtained during this program, the
standards were then set by multiplying these levels by 10 percent,
i.e., a 90 percent reduction. These numbers, once identified, then
became the required standards. The 10 percent AQL does not change
the values of the standards; it merely reflects EPA's view that
every production vehicle must comply with the established stan-
dards. This is consistent with EPA's finding, as discussed in
Section 3(a), that every production vehicle must comply with
standards established under Section 202 of the Clean Air Act.
Also, as explained in the legislative history of Section 202(a)(3)
(A)(ii), Congress intended that production line testing of heavy-
duty vehicles ensure that "each production vehicle will comply in
actual use." H.R. Rep. No. 95-294, 95th Cong., 1st Sess. 276
(1977). See also H.R. Rep. No. 95-564, 95th Cong., 1st Sess. 171
(1977). This language conclusively answers manufacturers arguments
that by use of the term "average" in Section 202(a)(3) (A) (ii),
Congress meant to reject the "every vehicle" concept.
-------
EPA has performed an analysis which indicates that a 10
percent AQL can cause a manufacturer to design to lower target
emission levels than those required by a 40 percent AQL. However,
the magnitude of the difference between the target levels depends
on several factors, some of which are within the manufacturer's
control. One of the most important of these factors is the vari-
ability of identical production engines (the "width" of emissions
distribution) at each design level. By increasing quality control
and minimizing other variations in the manufacturing and assembly
process, the manufacturer may reduce variability and raise the
target emission levels which it needs to meet. In practice, the
Agency believes that each manufacturer will trade off to one degree
or another lower design targets versus stepped-up quality control
to obtain the most cost-effective approach towards the 10 percent
AQL goal.
c. The Consistency of the 10 Percent AQL with the 40 Percent
AQL in Effect for the Current Light-Duty SEA Program
The 40 percent AQL was established for the current LDV SEA
program to implement the program in a manner not unreasonably
burdensome to the affected manufacturers. At the time the current
LDV SEA was proposed, several auto manufacturers stated that they
built the average production vehicle to meet the standards. It is
important to note that the situation at the time the 10 percent AQL
was proposed for the current LDV SEA program is different from the
situation today. As discussed in the Preamble to the current LDV
SEA regulations:
"The approach taken here, then, of not setting the AQL at 10%
will provide manufacturers the time and flexibility to bring all
their vehicles into conformance with the standards on a reasonable
schedule. Such a schedule can be compatible with their parallel
efforts to improve fuel economy and which does not expose them
unduly to the risk of loss of certification while they are learning
to bring their production vehicles into compliance with the law."
(41 FR 31475, July 28, 1976).
The circumstances under which the proposed LDT SEA program is
being promulgated are significantly different than those under
which the current LDV SEA program was promulgated. When the LDV
SEA program was promulgated the emission standards were already
established for then-current and future model years and manufac-
turers were building vehicles to meet the standards on the aver-
age. The short notice that manufacturers were given before the
proposed implementation of a 10 percent AQL .for the LDV SEA program
would not have provided sufficient time to make the necessary
design and production changes, thus causing a severe economic
impact on them. The proposed LDT SEA program will not be in
effect until 2 1/2 years in the future; EPA has determined that
this is this is sufficient leadtime to comply in a cost-effective
manner with all of the 1983 regulatory requirements, which include
the 10 percent AQL. (See Section E of the Summary and Analysis of
I GO
-------
Comments for a discussion of "leadtime.") The LOT industry was put
on notice in 1976 that EPA intended all motor vehicles to comply
with the standards when the Agency made the above statement in the
Preamble to the current LDV SEA regulations. When the proposed LDT
regulations go into effect in 1983, the industry will have had
seven years to develop the quality control procedures and institute
the design changes necessary to meet a 10 percent AQL.
EPA's approach in the current LDV SEA program, the HDE SEA
program and the proposed LDT SEA program is a consistent one: The
Agency has endeavored to implement an SEA program consistent with
its legal interpretation that every vehicle or engine must meet
standards and in a practical manner that does not place an unfair
or unreasonable economic or technological burden on the affected
industry. In the LDT case, the Agency has determined that the
final standards, in combination with a 10 percent AQL, are tech-
nologically attainable and can be implemented in a cost-effective
manner (see Chapter VII of the Regulatory Analysis).
d. Relationship of a 10 Percent AQL Program to Certification
Requirements
Several commenters indicated that they felt the present
certification program embodied an averaging concept which con-
flicted with the concept of a 10 percent AQL. They argued that
consistency required use of a 40 percent AQL so that essentially
the average engine emission level would meet thfe standards.
The staff does not agree with this contention. Section
206 of the Clean Air Act authorizes a certification program
(206(a)) and an assembly line testing program (206(b)). If a new
motor vehicle or engine design demonstrates compliance with Section
202 standards throughout its useful life, a certificate of con-
formity will be issued under 206(a) regulations. The certificate
is issued with respect to Section 202 regulations, i.e., regula-
tions establishing emission standards. Since the function of the
assembly line testing program is "to determine whether new motor
vehicles or engines being manufactured do in fact conform with
regulations with respect to which the certificate of conformity was
issued," the program will determine compliance with emission
standards.
The EPA certification and SEA programs attempt to accomplish
different but related objectives. Because of the differences, the
programs need not necessarily employ the same approaches towards
compliance. Through certification, a manufacturer demonstrates
that it has the capability to design a vehicle or engine that will
comply with emission standards promulgated under §202(a) through-
out its useful life under conditions simulating actual use. Once
these prototype vehicles or engines demonstrate compliance, EPA
issues the manufacturer a certificate of conformity allowing it to
actually manufacture vehicles or engines similar to the prototypes
for distribution into commerce. Then SEA requires the manufacturer
-------
to demonstrate that newly manufactured vehicles or engines will
also comply with standards throughout their useful lives. As
discussed in section a, EPA has determined that every production
vehicle or engine must be in compliance.
e. Cost Impact of the 10 Percent AQL on Light-Duty Truck
Manufacturers
There is a cost component attributable to the 10 percent AQL,
as there is to all other compliance options in the regulatory
package. A light-duty truck manufacturer will actually incur a
"10 percent AQL" cost in those cases where it experiences diffi-
culty in attaining the target emission levels, i.e., when the
manufacturer must spend more money in going to the 10 percent AQL
target level from some other (higher) level, or where it decides to
step up its in-house quality control programs in response to a 10
percent AQL.
A cost-effectiveness analysis has been performed in conjunc-
tion with the evaluation of this regulation. One option examined
was the cost of the proposed SEA program at <± 40 percent AQL versus
its cost at a 10 percent AQL. The analysis indicated that the 10
percent AQL SEA program is the more expensive option, but that the
cost of moving to the 10 percent AQL is small relative to a 40
percent AQL.
f. Air Quality Impact of a 10 Percent AQL
EPA has performed an analysis of the reduction in emissions to
be obtained in going from a 40 percent AQL to a 10 percent AQL in
the SEA program. This analysis appears in Chapter VII of the
Regulatory Analysis. The findings of this analysis indicate that
by implementing a 10 percent AQL LDT HC emissions will be reduced
an average of .006 tons per vehicle over the vehicle's lifetime,
LDT CO emissions will be reduced 0.2 tons, and LDT NOx emissions
will be reduced 0.04 tons.
As shown on Table VII-A in the Regulatory Analysis, these
reductions represent a positive reduction in HC, CO and NOx for
LDTs which EPA analyses have shown can be achieved in a cost-
effective manner. On the basis of dollars spent per ton of emis-
sions removed, the 10 percent AQL compares favorably with other
emission control strategies.
I
g. -The Effect of the 10 Percent AQL on Fuel Economy / I
;
In order to go from meeting the proposed 1983 LDT standards at
a 40 percent AQL to meeting these standards at a 10 percent AQL it
will be necessary to : 1) upgrade the air injection system of one
LDT, 2) add EGR to a diesel LDT configuration, 3) recalibrate the
EGR on some gasoline LDTs, 4) change the engine calibration for
some LDTs and 5) increase the average catalyst loading slightly.
(See the discussion of the 10 Percent AQL Compliance Costs in the
-------
will involve a small fuel economy penalty. When estimating the
fuel economy impact of this package, EPA determined that if
the manufacturers ignored fuel economy considerations in design and
introduced no new technologies, the overall fleet fuel economy
potential will be reduced less than 4 percent. The 10 percent AQL
would account for a negligible part of the fuel economy penalty
under this scenario. EPA believes that LDT manufacturers can avoid
a fuel economy penalty entirely by using electronic engine con-
trols. (See the discussion on Fuel Economy in the Summary and
Analysis of Comments.) Electronic engine controls are already
being used in 1980 passenger cars, and EPA believes that in 1985
most LDT manufacturers will use them to meet the tighter NOx and
fuel economy standards.
4. Staff Recommendations
It is recommended that a 10 percent AQL be promulgated
in the Final Rule. An SEA program with a 10 percent AQL is
consistent with EPA's legal interpretation of the Clean Air
Act, does not place unreasonable cost burdens on light-duty
truck manufacturers, results in a positive reduction in emissions,
has no significant impact on fuel economy, and is technologically
feasible, given the emission standards to be promulgated.
Other Selective Enforcement Auditing Issues
1. Definition of "Configuration" (§86.1002(b)).
EPA proposed that a light-duty truck (LDT) configuration
be "...described on the basis of... other parameters which may be
designated by the Administrator." GM contested this definition as
being unreasonably broad and vague and wanted protection against
arbitrary selection of parameters by EPA.
This provision about "other parameters" is similar to a
provision contained in the present light-duty vehicle (LDV) SEA
definition of "configuration". A LDV configuration has never been
defined beyond the specific parameters contained in that defini-
tion. Present LDT configurations can be described using the
specific parameters in the LDT definition. However, EPA needs some
flexibility in specifying configurations, because new emission
control technologies developed in response to 1983 and later LDT
standards may result in emission control parameters not presently
identified. EPA does not intend to use this flexibility in an
arbitrary manner but has retained the proposed definition in the
final rule.
2. Low volume LDT families are not exempt from SEA test
orders (§86.1003).~~
This exemption was not included in the proposal. MVMA
recommended that if the "projected sales volume for a given
101
-------
code is less than 2,000 trucks for the current model year the
manufacturer should be permitted, upon receiving a test order, to
petition the Administrator for an exemption from Selective Enforce-
ment Auditing of that engine code." Section 206(b) of the CAA does
not exclude any particular class or category of production vehicles
or production engines from its provisions. Therefore, from a legal
standpoint, all production LDTs are potentially subject to testing
to determine compliance with applicable emission standards. From a
practical standpoint, to the extent that test orders for low volume
configurations are issued, EPA believes that its newly developed
sequential sampling plans allow these configurations to be tested
as expeditiously as possible because, with these plans, as few as 7
LDTs are required to pass an audit, and the LDTs to be tested may
even be selected over several days. Therefore, the impact on
customer delivery schedules should be minimized. Because the
sequential plans allow the flexibility to deal with low-volume
configurations, no change has been made in the final rule.
3. Statement about "...instructions in the test order1'
is not redundant and unnecessary (§86.1003(b)).
GM stated that this phrase, in the last sentence of paragraph
(b), should be eliminated because the CAA mandates compliance with
test orders issued under regulations. EPA prefers not to delete
the phrase because it alerts the manufacturer of its obligations
directly in the regulations under Subpart K.
4. Vehicle selection procedures in the test order (§86.1003-
83(c)(D).
The proposal stated that "the test order will specify...
the procedure by which LDTs of the specified configuration must be
selected." GM stated that this statement was vague and ambiguous.
Both GM and MVMA said that this allowed too much flexibility and
could impede the expeditiousness of the audit and that by not
including these procedures in the regulations EPA was denying the
LOT manufacturers the right to comment on them.
EPA has made no changes in its proposed statement for the
final rule. Commenters stated that if procedures can be stan-
dardized, they should be placed in the regulation. However, it is
not possible to standardize the selection process because of the
different production procedures encountered at the assembly
plants. Neither GM nor MVMA offered any useful suggestions as to
how to standardize selection techniques. In addition, EPA's
experience with selection, during the current SEA program, indi-
cates that it is necessary to have flexibility in the selection
process to account for the unique situations encountered during
vehicle selection at different assembly plants. EPA intends to use
the flexibility to expedite vehicle selection and therefore does
not believe that this will impede the progress of audits. It
should be emphasized that paragraph §86.1007-83(a) allows for
manufacturer input into the determination of the selection pro-
cedure.
-------
5. Other standardized test order instructions (§86.1003-83
~ ~
EPA proposed that "In addition, the test order may include
other directions or information essential to the administration of
the required testing." GM stated that this statement was vague and
ambiguous: "Any procedures which can be standardized should be
placed in the regulations and information deemed to be "essential"
should also be included so that the manufacturers are not effec-
tively precluded from commenting on such information or pro-
cedures ."
EPA has determined that some of the specific instructions
presently incorporated in LDV SEA test orders are applicable
to LOT SEA testing and has included them in the final rule as new
paragraph §86. 1003-83(c) (2) . As the need for new instructions
which can be standardized becomes apparent in the future we intend
to amend the regulations to incorporate them. However, the pro-
vision to include "other directions or information" essential to
administer SEA testing has been retained to allow some flexibility
in SEA operating procedures. This flexibility can be in the
interest of the manufacturers and EPA, as it will allow audits to
be conducted in the most expeditious manner practical, given
circumstances unique to a particular manufacturer.
The latitude built into the test order and sample selection
sections of the SEA regulations is intended to accomodate pro-
cedural variations, especially in the area of LDT selection.
Specific instructions may be made to minimize the impact on
each manufacturer's normal production activities while still
assuring the generation of accurate, representative test results.
6. Selection at non-preferred plants (§86.1003(d)K
EPA proposed that, even though a manufacturer has submitted a
list of assembly plants preferred for LDT selection, "the Admini-
strator may order testing at other than a preferred plant." GM and
MVMA believed that manufacturers should be able to designate from
which plant a specific configuration will be selected. GM stated
that it was necessary for the manufacturers to designate those
plants in order to ensure that vehicle selection does not disrupt
normal production practices at some plants and to allow the
manufacturers to have an input into the selection process.
The sequential sampling plans contained in this regulation
were designed to allow flexibility in sample selection to prevent,
to the greatest extent possible, disrupting a manufacturer's normal
production and delivery schedules. EPA intends to select the
sample of LDTs at preferred locations, but requires the flexibility
of selecting at non-preferred plants when that would allow the
audit to be performed more expeditiously or permit the auditing,
based upon available evidence, of specific cases of noncompliance.
To retain this flexibility, EPA made no change to the final rule.
-------
7. Clarification of the projected annual sales used in
determining the annual limit on test orders. (§§86.1003
(f)(l)(i) and 86.1003(f )(l)
EPA proposed that the annual limit of SEA test orders be
calculated using the projected "sales for that year." Although no
comments were received on this issue these paragraphs have been
revised to specify that the annual limit is calculated using the
projected "sales bound for the United States market for that year"
in order to clarify the intent of these paragraphs and to make them
consistent with the previously promulgated heavy-duty engine
regulations (44 FR 9488, February 13, 1979).
8. Test orders that will count against the annual limit
(§86.1003-83(f)(2), (3)).
EPA proposed that test orders will not count against the
annual limit under the following circumstances:
(1) The configuration being tested fails according to the
sampling plan decision criteria;
(2) Testing is not completed;
(3) The test order is issued on the basis of any evidence
indicating noncompliance with the AQL; and
(4) Follow-up audit testing is conducted on a configuration
which previously had its certificate of conformity
suspended or revoked.
GM, Chrysler and MVMA stated that failed test orders which pass
a follow-up audit and test orders issued on the basis of evidence
of noncompliance should count. They said if these test orders did
not count, there would not be an upper limit and manufacturers
could not plan the extent to which facilities must be allocated to
meet this requirement.
Because of its responsibility to investigate those LDT
configurations for which it has evidence of noncompliance, EPA will
not establish an absolute limit on the number of test orders it
will issue. However, EPA is sensitive to the manufacturers'
concerns that they may be subjected to an indefinite number of test
orders. Therefore, the proposal has been amended to provide that,
when based on evidence of noncompliance, a test order issued within
the annual limit will count toward the annual limit, if the' con-
figuration passes the audit. If the limit has been reached,
additional test orders can be issued only on the basis of evidence
of noncompliance. In addition, the provision requiring a statement
of the reason for issuance of a test order beyond the annual limit
will be retained.
Follow-up audits do not count toward the annual limit because
106
-------
they fall under the "umbrella" of the original test order, which
resulted in a fail decision; see paragraph §§86.1012-83(j)(2)
and (k)(2). The only exception to this provision is when a
configuration, having had its certificate of conformity suspended
or revoked by EPA, is then "replaced" by another configuration that
was previously certified.
9. Selection of test vehicles by the Administrator (§86.1004
(a)).
EPA proposed that when SEA testing will be performed by
the Administrator, he will select LDTs "...in a manner desig-
nated by him..." GM suggests that this provision be revised so
that LDTs are selected in accordance with §86.1007-83. EPA
intends to select LDTs for its own testing in a manner consistent
with the standardized selection procedures established in con-
sultation with that manufacturer'. The final rule therefore
includes the statement that the Administrator will select his
test LDTs "...in a manner consistent with the requirements of
§86.1007-83..."
10. Discrepancies between EPA test results and manufacturer
test results. (§86.1004(b), (c)).
EPA proposed that its test results comprise the official
data for a test vehicle when there is a disagreement with the
manufacturer's results. GM disagrees with the assumption that
the manufacturer's test facility is deficient and that it bears
the burden of proving that its own data is correct. However,
the regulations provide two mechanisms for resolving differences
between data: (1) paragraph § 86.1004-83(c)(2) allows a manu-
facturer to demonstrate that EPA's data were erroneous and its
own data was correct; and (2) if EPA invokes a suspension of
the certificate of conformity based on the Administrator's test
data, the manufacturer can request a hearing under paragraph
§86.1012-83 (1) to determine whether the tests were conducted
properly. Therefore, EPA is not changing this provision.
I
11. Retaining names of personnel (§86.1005-83(a)(2)(iii),
TTvTT:
EPA proposed that the manufacturers be required to retain the
names of all personnel involved in the conduct of the test and in
the supervising and performance of a repair. GM objected to these
requirements, stating that the information "is unnecessary and
irrelevant for EPA's needs and goes beyond that required by current
LDV and LDT regulations."
EPA believes that the names of manufacturer personnel involved
in SEA audits should be available if an investigation of the
conduct of an audit is necessary. However, EPA does agree with GM
that it is unnecessary to include this information in the audit
-------
suggested by GM, but paragraph §86.1005-83(a)(2)(ii) in the final
rule requires the manufacturer to retain the names of all personnel
involved in the audit for future reference.
12. Requirement for submitting manufacturer's test results
(§86.1005(c)).
Several manufacturers were critical of the proposed provision
requiring them to submit their own production LDT test data. GM,
Ford, Chrysler, Toyota and MVMA considered this requirement un-
reasonable and an unjustifiable burden. GM stated that these data
do not necessarily reflect overall production emissions performance
throughout the model year. GM and Toyota objected to EPA's use of
this data for enforcement purposes and along with Chrysler believed
that this is a deterrent for manufacturers to continue in-house
auditing. Chrysler believes that this requirement is not in
conformance with the Clean Air Act. Chrysler stated that "Congress
intended that the Administrator be only entitled to access to the
data he needs to determine compliance on a specific basis".
(Chrysler's emphasis) Ford and MVMA also believed that data submit-
ted to EPA should be limited. MVMA suggested that only data from
complete emission tests of production vehicles need be submitted.
Ford proposed that the manufacturers only be required to report the
results of regularly scheduled tests.
Subpart K does not impose any requirement that a manufacturer
conduct an internal quality audit program, but if it does conduct
this type of program, Section 208(a) authorizes the Administrator
to require the submission of this data because the data can help
determine compliance of LDTs with applicable emission standards.
This requirement has been proven workable in the LDV SEA program
and does not appear to be unreasonably burdensome to manufacturers.
EPA believes that the reporting requirements it proposed for
the LDT manufacturers are reasonable because these requirements
are similar in scope to those currently being met by LDV manu-
facturers. However, EPA has made two changes in response to
comments received: (1) The manufacturer is required to describe
the emission test used to obtain the data submitted; see §86.1005-
83(c)(l). This change will help EPA determine the degree of
correlation between a "short test" result, if the manufacturer
uses this kind of test, and full FTP data. (2) The manufacturer
need only submit data on ADP storage devices if these devices are
compatible with EPA equipment. EPA will furnish the necessary ADP
storage devices upon a manufacturer request.
13. Entry and access (§86.1006).
Ford contends that the provision allowing EPA inspectors
to "inspect and monitor any aspect of engine or vehicle test
procedures or activities1' is beyond EPA's authority in the CAA
because it would allow EPA to have access to those areas of manu-
facturing facilities which have no connection at all to a legiti-
mate EPA interest. Ford wants EPA to limit its inspections to
1C*
-------
those items specifically listed in Section 86.1006-83(b)(2) .
Section 206(c) of the CAA allows EPA "... (1) to enter, at
reasonable times, any plant or other establishment of such manu-
facturer, for the purpose of conducting tests of vehicles ... or
(2) to inspect, at reasonable times, records, files, papers,
processes, controls, and facilities used by such manufacturer in
conducting tests under regulations of the Administrator." Ford did
not explain how EPA was exceeding its authority in Section 206(c)
or any other section of the CAA. EPA must be able to monitor and
inspect any aspect of vehicle test procedures and activities
necessary to assure that these vehicles are being prepared and
tested according to applicable regulations and the prescribed
Federal Test Procedure. In addition, EPA at this time cannot
develop a specific comprehensive list of all test procedures or
activities which could require inspection or monitoring during the
course of an audit. If Ford belieVes that EPA is conducting an
inspection in excess of its authority, it may not consent to the
inspection. The Supreme Court decision in Marshall vs. Barlow's,
Inc.. has limited inspections without a search warrant to those
which have the manufacturer's consent. If voluntary consent is
refused, EPA will not attempt to enter any of a manufacturer's
facilities, including emission testing facilities, without first
obtaining a search warrant. GM suggests that only "emission
related" parts be investigated in paragraph §86.1006(b)(4), but
gave no reason for this comment. §86.1006(a) states that only
matters related to Subpart K will be investigated. EPA has not
revised §86.1006 in response to these comments.
14. Entry and access in foreign jurisdictions (§86.1006(g)).
EPA proposed that foreign testing and manufacturing facili-
ties must be located so as to permit EPA entry and access. VW
requested that this requirement be deleted because manufacturers
have no control over a country's policy.
EPA has made no change in its proposed statement for the final
rule. To allow a manufacturer to produce vehicles in a country
which did not permit EPA to conduct inspections would inhibit EPA's
ability to enforce air pollution regulations at that location.
Further, it would give that manufacturer an unfair advantage over
other manufacturers which are subject to EPA inspections. When a
manufacturer decides to locate an assembly plant in a foreign
country, one of the risks it is .voluntarily accepting is that the
country will continue to allow EPA inspectors to conduct assembly
plant inspections within that country. EPA must be able to conduct
inspections at these foreign assembly plants to accomplish the
tasks mandated by Congress.
15. Authorization for personnel appearances and entry
without 24 hours notice. (§86.10Q6(h)(4)).
/tf
-------
GM recommended that proposed paragraph §86 . 1006-83(h) ( 4)
be amended to require approval of the Assistant Administrator
for Enforcement before manufacturer personnel could be questioned
by EPA investigators. GM and VW also recommended adding a new
paragraph §86. 1006-83(h) (5) which would require the Assistant
Administrator to sign test orders in which manufacturers are not
given 24 hour notice before entry. EPA believes it is unnecessary
to require the Assistant Administrator to authorize either appear-
ances of personnel or entry without 24 hours notice because these
authorizations can be performed by other responsible Agency offi-
cials. If a manufacturer refuses to consent to personnel appear-
ances or entry without 24 hours notice, EPA is required to seek a
search warrant before attempting to conduct these activities.
Therefore, no changes relating to these issues have been made in
the final rule.
16. Order of test results (§86.1007(e)) .
EPA proposed that the test order will specify the order in
which test results will be used in applying the sampling plan.
This paragraph was also proposed in the rulemaking concerning
gaseous emission regulations for heavy-duty engines (HDEs) (44 FR
9488, February 13, 1979). Based upon a comment received during
that rulemaking, EPA changed the final version of that paragraph
(45 FR 4173, January 21, 1980). The revised paragraph specifies
that the order of sample selection determines the order of the
decision sequence. Although no specific comments were received
relating to this paragraph during the 1983 LDT rulemaking, EPA has
revised this paragraph, to be consistent with the HDE regulations,
for purposes of uniformity and clarification. This will provide a
consistent basis for applying test results to the sequential
decision criteria.
17. Retention and shipment of test vehicles (§86.1007(f)) .
EPA proposed that all untested LDTs in the test sample
be kept on hand until a pass or fail decision is reached. GM
indicated this might involve the selection of the maximum number of
test vehicles before testing began and their retention until the
audit was completed. Under the sequential sampling plan, vehicles
may be selected as required to assure expeditious testing. The
maximum number of vehicles associated with a sampling plan need not
be selected prior to initiating testing. Further, any LDT which
has passed the required emission test during the SEA audit may be
shipped. However, once a manufacturer ships any LDT from the test
sample, as defined in §86.1002-83, it relinquishes its prerogative
to conduct retests during that audit.
18. Allowance for a "dealer preparation" procedure (§86.1008
GM, Ford, MVMA, Chrysler and AMC believe EPA should allow the
-------
manufacturers to conduct a predelivery inspection on SEA vehicles.
GM, Chrysler and Ford believe this inspection is necessary to
assure that the results of SEA testing are representative of the
emission levels of vehicles on the road. GM states that disal-
lowing the dealer preparation increases the stringency of the
emission standards because EPA would require the manufacturers to
build cars which meet the deteriorated emission standards off the
assembly line. GM also believes that this action is inconsistent
with §85.2108 of the recent NPRM on 207(b) Warranty Regulations.
AMC stated that disallowing the predelivery inspection would
force them to institute extra inspections on the assembly line.
These extra inspections would be redundant because they are per-
formed by the dealers.
EPA believes that SEA test vehicles which have undergone
dealer preparation procedures will represent "real world" condi-
tions only to the extent that these procedures are actually and
correctly performed by dealers. §86.1008(b)(l) of the proposed
regulations does permit a dealer preparation procedure to be
performed if it is approved in advance by the EPA Administrator.
§85.2108 of the proposed Warranty Regulations requires dealers to
furnish the purchaser of a new light-duty motor vehicle with a
certificate which states that the vehicle meets emission standards
To make this certification in good faith the dealer would have to
perform all emission related preparation required by the manu-
facturer. The certificate in itself does not prove that the dealer
preparation procedures are being performed properly in the "real
world". EPA's experience with light-duty vehicles (LDVs) indicates
that in several cases, dealer preparations are not performed, or
are performed incorrectly. For these reasons, unless the manu-
facturer can demonstrate to the Administrator that these dealer
preparation procedures are carried out routinely and correctly at
the dealerships, the Administrator may require additional infor-
mation, such as dealer survey data, which demonstrates that dealer-
ships are performing the dealer preparation correctly before
allowing a manufacturer to perform certain dealer preparation
procedures during SEAs.
I
19. Time allowed to ship test vehicles (§86.1008(e)).
EPA proposed that the manufacturers be required to ship
test vehicles to the testing location within 24 hours of selection,
unless the Administrator approves a greater shipping time based on
a request from the manufacturer. GM believes that the 24 hour
shipping time is not applicable to low volume configurations and
requests that a provision allowing a greater period of time to ship
these vehicles be incorporated into the regulations.
EPA has made no change in the final rule because EPA may
approve a different time period based on a satisfactory manu-
facturer request. If GM can adequately justify more time for the
situation it mentions a time extension may be granted.
-------
20. Option to retest (§86.1008(1)).
GM requested that EPA revise this proposed paragraph to allow
retesting at any time during the audit (as opposed to only after a
fail decision has been reached) and to delete the requirement for
testing each LDT the same number of times. GM claims that these
changes are necessary in order to minimize the possible logistic,
storage, and economic impacts upon a manufacturer's operations.
Selective retesting of test vehicles could bias the test
results from a statistical viewpoint, because vehicles not retested
may produce different results, in terms of pass or fail, than they
did in the first test. Therefore, EPA will require that each LDT
be tested the same number of times. EPA realizes that there may be
certain situations in which retesting prior to reaching an audit
decision will require fewer emission tests to complete the audit.
Therefore, EPA has revised §86.1008(i) to allow for retesting
before an audit decision is reached, upon approval by EPA of a
manufacturer's justified request to retest.
This provision is also being incorporated into the previously
promulgated heavy-duty engine regulations for consistency with the
LDT approach. Since EPA considers this a relaxation of the previ-
ously promulgated heavy-duty engine provision, no reproposal of
this paragraph is being made.
21. Rounding of final test results (§86.1009(b)).
GM recommended that the final test results be rounded to
3 decimal places, instead of the proposed 2 decimal places,
to provide for a more accurate computation. GM did not provide any
analysis to support its recommendation. Upon review of this issue,
EPA has revised the method of calculating the final deteriorated
test results in SEA to make it compatible with the method used
during certification, which is based on the number of significant
figures instead of decimal places. To calculate the final deteri-
orated test results, the actual emission results are rounded to the
same number of decimal places contained in the applicable standard
expressed to 1 additional significant figure. This number is then
multiplied by the DF and rounded to the same number of significant
figures contained in the standard. This method will allow for a
direct comparison between the final deteriorated test results and
the standard. This change has also been incorporated into the
previously promulgated heavy-duty engine regulations for purposes
of generalizing the rounding procedure for all standards. '
/ /
22. . Calculation of final deteriorated test results (§86.1009
(c)(l) and (c)(2)).
GM believes that these proposed paragraphs should be revised
to eliminate the inconsistency in proposed §86.1009 between para-
graph (c) and (c)(2). These paragraphs have been amended to remove
the inconsistency noted by GM and to clarify the use of DFs. In
-------
GM's suggested revision of the proposed paragraphs no mention was
made of adjustment of DFs according to certification conventions.
Multiplicative DFs which were determined during certification to be
less than one are set equal to one. Similarly, additive DFs which
were less than zero during certification are set equal to zero.
These DF conventions permit comparisons to be made between certifi-
cation and SEA test results.
23. Reporting of Invalid Tests (§86.1009(d)(5)(iii).
GM objects to the proposed provision that requires the
manufacturers to report the test result of invalid tests and
the reason for invalidation. GM states that this information
is not required by the Administrator to determine whether a
manufacturer is acting in compliance with the regulations.
EPA believes that this information is necessary for the
Agency to determine if the decision to invalidate the test was in
compliance with the regulations. The results of all emission tests
performed during an SEA must be available in the event that the
Agency subsequently declares valid an emission test the manu-
facturer had declared invalid. For these reasons no change has
been made to the proposed provisions.
24. Use of test results following a pass or fail decision for
a particular pollutant (§86.1010(c)).
GM suggests that this paragraph be modified by adding a
sentence to indicate that once the test sample is accepted or
rejected for a particular pollutant, additional test results for
the pollutant will not be considered for auditing purposes.
EPA. has added a clarification to the final rule in essentially
the form proposed by GM, except to change the wording to read,
"Once a pass or fail decision has been made for a particular
pollutant..." to retain consistency with sequential sampling plan
terminology.
25. Batch Sampling vs. Sequential Sampling for Vehicle
Selection (§86.1007)"
VW encouraged EPA to continue use of batch sampling plans,
because VW believed an SEA decision under the sequential sampling
plan could be made based on results from unrepresentative vehicles.
The manufacturer stated that "the proposed sampling technique no
longer ensures that sample vehicles are drawn all across the
manufacturer's production output." VW contended that since an
audit could consist of one single sample drawn at one time, a
manufacturer could fail an audit on the basis of a quality control
problem encountered on a single shift.
EPA is implementing sequential sampling plans instead of batch
-------
type sampling plans because the sequential plans often require
fewer tests to reach an audit decision than the batch plans, fewer
vehicles need be retained during the audit, and the simpler deci-
sion rules ease administration. When conducting an SEA of high
volume configurations under the current batch sampling plan,
several batches of vehicles may be selected over one shift. The
decision rules for batch sampling allow for audit failure after two
batches. Therefore, the risk of failing an SEA based on a single
shift's production also exists under a batch sampling plan.
If a manufacturer believes it has failed an audit due to
quality control problems, the regulations provide a mechanism for
the manufacturer to prove its contention. §86.1012(j) allows the
Administrator to reinstate the suspended certificate of conformity
when the manufacturer: 1) submits a written report which identi-
fies the reason for noncompliance and describes the proposed remedy
and 2) demonstrates that the vehicle configuration now complies
with the standards. Under this provision, if a manufacturer
believed it failed an SEA because of a problem on the assembly line
during a single shift, the manufacturer could have its certificate
reinstated after it submitted a report describing this problem,
demonstrated to the Administrator that steps had been taken to
remedy the problem, and showed that the vehicles now comply with
the standards through a reaudit.
26. Complexity of the Sampling Plan (§86.1010(c)).
Ford concurs with the proposed sequential sampling plan
for SEA test purposes; however, "Ford is still dismayed at the
unnecessary complexity of the plan, as proposed, with its various
code letters, annual sales, and phases." Ford developed a single
sampling plan with a 40 percent AQL which could be used for all
engine families with yearly projected sales above 50.
Ford did not explain why the proposed sampling plans are
unnecessarily complex. The decision as to which sampling plan to
use for a particular configuration is dictated by the projected
annual sales of that configuration. Manufacturers supply this
information to EPA in their Application for Certification, and
therefore the information is readily available to both EPA and the
manufacturer. EPA does not believe its proposed sampling plans are
overly complicated.
The sampling plan that Ford developed incorporates a 40%,AQL.
EPA plans to implement a 10% AQL for LDT. Further, Ford's sampling
plan requires the manufacturers to do more testing than would be
necessary using the sampling plans proposed by EPA. EPA believes
that it is desirable to reduce the number of tests a manufacturer
must perform during an SEA to a minimum. EPA has therefore not
revised §86.1010(c) in response to this comment.
27. Application of suspension decisions to all manufacturing
plants (§86.1012(6)7!
Hf
-------
EPA proposed that a suspension decision imposed on LDTs of a
failed configuration produced at one plant could also be imposed on
LDTs of the same configuration produced at all other plants.
Chrysler, VW and MVMA stated that the proposed regulations did not
allow for a partial suspension or revocation. Chrysler and MVMA
requested that EPA allow the manufacturer « chance to demonstrate
that a problem was plant specific before the certificate was
suspended or revoked for that configuration.
Paragraph §86.1012-83(e) states,11. . .the Administrator may
suspend the certificate... at all other plants" (emphasis added).
The^use of the word "may" indicates that this is a discretionary
decision on EPA's part and that the suspension order will not
automatically apply to all other plants. Any manufacturer has an
opportunity, as provided in Subpart K, to demonstrate that the
suspension order should not apply to other plants. §86.1012(1)
allows a manufacturer to request a'hearing for a suspension under
§86.1012(e). §86.1014(a) states that hearings are applicable to
requests under §86.1012(1). S86.10L4(c)(2)(ii)(B) states, in
pertinent part, that when a. hearing is requested under §86.1012(1),
the issues will be restricted to proper conduct of tests and
proper application of sampling plans, specifically,"...whether
there exists a basis for distinguishing LDTs produced at plants
other than the one from which LDTs were selected for testing which
would invalidate the Administrator's decision under §86.1012(e)".
EPA has therefore made no change to this paragraph in the final
rule.
28. Failed vehicle report (§86.1012(i)(2)).
EPA proposed that the report on corrective testing of those
vehicles that failed emission testing during an SEA be submitted to
EPA "...within 5 days after, completion of testing..."- GM proposed
that the 5 day requirement be deleted in order to provide a reason-
able period of time to the manufacturers and for comparability with
the LDV SEA requirements. To clarify its intention, EPA has
revised this paragraph for the final rule to read "...within five
working days after successful completion of testing on the failed
engine or vehicle...". While EPA needs to receive reports on the
repair of noncomplying LDTs in a timely fashion, corrective action
need not be taken immediately after a failure, so long as the
manufacturer does not attempt to introduce a failed LOT into
commerce.
29. Qualifications of a Judicial Officer (§86.1014(b)(5)
(iii)).
GM proposed that this paragraph be amended to ensure that the
Judicial Officer (J.O.) is a graduate of an accredited law school
and a member of a bar association.
EPA does not take issue with GM's proposal and has incor-
porated it into the final rule.
UV A
-------
I. Issue: Nonconformance Penalty
This issue concerns the system for production compliance
auditing (PCA) and nonconformance penalties (NCP) proposed in the
NPRM. Since the proposed emission standards were considered
feasible for all manufacturers to meet, nonconformance penalties
were not made available in the NPRM.
As described elsewhere in this Summary and Analysis of Com-
ments (see G. "Technological Feasibility"), the EPA staff still
believes that the standards are attainable by all manufacturers.
However, EPA has decided not to finalize nonconformance penalties
at this time. The possible need for making nonconformance penal-
ties available and their role in the compliance process is still
under review by the Agency. These matters will be the subject of
separate rulemaking by EPA. Therefore, comments received on these
aspects of the NPHM will not be dealt with at this time.
Delaying final action on PCA/NCP does not adversely affect the
mnufacturers' ability to comply with the regulations finalized in
this present rulemaking. With or without PCA/NCP, it is EPA's
intent for all manufacturers to comply with the standards. NCPs
are not to be viewed as a route to a less stringent emission
standard via intentional design to a higher emission rate and
payment of the associated NCP.
-------
J. Issue: Diesel Crankcase Emissions Control
1. Summary of the Issue
The proposed regulations require that no crankcase emissions
shall be discharged into the ambient atmosphere from 1983 model
year (and later) light-duty truck diesel engines.
2. Summary of the Comments
The proposed crankcase controls for diesel LDTs drew con-
siderable adverse reaction. Both EPA1* justification and feasibil-
ity issues were addressed by most of the commenters.
Several comments pointed to the low hydrocarbon and carbon
monoxide emissions from light-duty diesel crankcases as evidence of
a lack of need for controls. Additionally, commenters felt that
the information quoted in the NPRM is inconclusive in establishing
the presence of nitrosamines in diesel blowby emissions.
The feasibility of controlling the crankcase emissions was
challenged on the basis that some light-duty diesel engines will be
equipped with turbochargers. The anticipated technical problems
arise from the oily nature of the blowby emissions which in a
simple system would be introduced into the inlet air supply.
Although in a naturally-aspirated engine the slight negative
pressure of the manifold can draw crankcase fumes into the combus-
tion chamber, the manifold of a turbocharged engine is under
greater pressure than the crankcase. Thus, unless it is pres-
surized, the blowby must enter the stream on the inlet side of the
turbocharger, allowing the oily emissions to become deposited on
the compressor wheel. It was indicated that this will result in a
decrease in turbo-efficiency which can detrimentally affect perfor-
mance, fuel consumption, and emissions.
Cummins Engine Company mentioned four means of crankcase
control which may have potential. None are developed to the extent
of assessing their feasibility. These four alternatives are:
I
1) Duct gases to turbo-inlet by way of a pressure regulator
and oil separator.
2) Draw gases through the regulator and separator and a pump
to the manifold downstream of the turbo (an expensive alternative
which still requires much development work to ascertain whether it
is satisfactory).
3) Aspirate and mix gases into the exhaust flow.
4) Pump the gas through the regulator and a separator into
the exhaust stream.
-------
3. Analysis of the Comments
All but a fraction of the light-duty diesel engines currently
being produced (for both LDVs and LDTs) have crankcase controls;
however, such control is not required by regulation. Regulation
was proposed in order that crankcase control be maintained in the
future and so that parity would exist between the diesel and
gasoline engines. Although the number of diesel LDTs is small at
this time, it is anticipated that the number will grow substan-
tially with time.
The majority of commenters saw no major problems associated
with the control of crankcase emissions in naturally-aspirated
diesel engines. The turbocharged engine, however, was quite
another matter. EPA received a number of comments essentially
stating that there are technical problems associated with crankcase
emission control of the turbocharged diesel, and that overcoming
these problems could be relatively costly. A possible solution was
recommended by Cummins (Suggestion #2 in Summary of Comments) and
seems to be worthy of investigation. By allowing the turbocharger
to be bypassed, the oil separator/pump/pressure regulator config-
urations would eliminate the excessive deterioration of the compo-
nent efficiencies. The pump itself might be affected to some
degree by the oily emissions; therefore, further analysis would
have to be performed.
Some commenters suggested that ducting the gases to the
turbo-inlet by way of a pressure regulator and oil separator
(Cummins Suggestion #1) may be the easiest and least costly of the
possible crankcase control methods. It was noted by the commenters
that even though this method seems viable there are problems
related to the oily emissions which can deposit and build up on the
compressor wheel causing a reduction in turbo-efficiency, fuel
economy, and durability.
Contrary to the conceptual arguments of the manufacturers,
however, is a concrete example of crankcase control of a production
turbocharged diesel engine. Mercedes-Benz, for some time now,
has been producing a light-duty diesel vehicle that is equipped
with a turbocharger and crankcase emission controls. In their
design, the engine blowby gases and the crankcase vapors flow
through a vapor line in the cylinder head to a cyclone separator
located in the air filter. The gaseous content of the crankcase
emissions is drawn off the separator by the intake manifold vacuum;
the fluid content is routed back to the oil pan. In an effort ito
gain further information regarding their design (essentially
Suggestion #1 by Cummins) the Staff at ECTD made telephone contact
with Mercedes engineers. The staff learned during the conversation
that Mercedes has had no problems to date with their deisgn and in
addition did not foresee any further problems such as significant
drops in turbocharger efficiencies. Obviously, such a method
must be practical and not be overly burdensome since it is being
-------
done without imposing regulations. We realize that there may
be differences between light-duty truck and light-duty vehicle
diesel engines, but the staff believes these differences to be few
and relatively minor. Therefore, the staff must conclude that the
proposed diesel crankcase emission control standard is technically
and economically feasible.
4. Staff Recommendation
The staff, after careful consideration, recommends that the
proposed crankcase emission standard for diesel light-duty trucks
remain unchanged.
-------
K. Issue: Numerical Standards/Standards Derivation
1. Summary of the Issue
EPA has proposed new emission standards for light-duty trucks:
0.8 g/mile HC
10 g/mile CO
These standards are derived from emission tests of uncontrolled
1969 model year light-duty trucks. These emission tests data were
used to calculate a baseline from which the proposed HC and CO
standards were derived. The 1969 Light-Duty Truck Test Procedure
(Title 40, CFR, Part 86, Subpart B) was used to obtain the emis-
sions data.
EPA is proposing that the current NOx standard of 2.3 g/mile
be retained for 1983 and later model years.
2. Summary of the Comments
The majority of the comments were directed at EPA's develop-
ment of emission standards in Technical Report SDSB-79-23 entitled
"1969 Light-Duty Truck Baseline Program and 1983 Emission Standards
Development. Commenters contend that improper mathematical and
engineering practices were employed to develop the LDT baseline and
proposed emission standards. Specifically, criticism of the test
program and standards derivation were:
a. Eighteen vehicles were too small a sample.
b. Vehicle stratification was inadequate. Only 1 Chevy 307
was tested, while 4 Chevy 350's were tested.
c. The vehicles were unrepresentative since all vehicles
came from San Antonio, Texas area and had mileages much less than
new proposed useful life and represented only 84 percent of the LDT
market.
d. The pre-test vehicle maintenance was improper for estab-
lishing the baseline.
e. Deterioration factors should have been calculated and
applied to the baseline and standards.
f. Sampling error should have been recognized in calculating
the standards. , ,
/
g- Standards are more stringent as a result of other .ele-
ments of the NPRM such as allowable maintenance and durability.
3. Analysis of the Comments
This section is divided into two Subparts. First, EPA's
discussion relating to the final 1969 LDT baseline emissions
-------
results is presented. Second, the comments on the issue are
analyzed.
a- Final 1969 LPT Baseline Emission Results
This section updates and summarizes the final 1969 LDT base-
line emission results. These results were obtained under EPA
Contract No. 68-03-2683 with EG&G, Inc. The initial baseline
results which appeared in SDSB Technical Report No. 79-23, "1969
Light-Duty Truck Baseline Program and 1983 Emissions Standards
Development", July 1979, are shown in Table 1.
These baseline results were used to calculate the light-duty
truck HC and CO standards, proposed on July 12, 1979, in FR, Vol.
44, No. 135. Subsequently, EG&G discovered that some of the
vehicles were tested without having had all the carburetor main-
tenance performed (letter of 9/8/79, EG&G to Mr. Larry Ragsdale).
As a result, all vehicles were rechecked and proper carburetor
adjustments were made to the vehicles where necessary. Any vehic-
les that needed adjustment were retes'ted.
Table 2 shows the final corrected emission test results for
the 18 vehicles included in the baseline report cited above.
Ten vehicles of the eighteen used to calculate the proposed
1983 light-duty truck HC and CO standards were retested. Addi-
tionally, Table 3 summarizes emission data for all 1969 light-
duty trucks tested under EPA's baseline program. Included in this
table are emission results from an additional three vehicles (#617,
#621, #623) that were not part of the original sample used to
determine 1969 light-duty truck baseline levels (refer to Table
1). These additional vehicles were tested after the 1969 baseline
levels had been determined and included in the 1983 Light-Duty
Truck NPRM.
Table 4 summarizes the sales-weighted emission results derived
from the various test samples (Tables 1, 2, and 3). Table 5
presents the emission levels that represent a 90 percent reduction
from the baseline levels in Table 4. Clearly, the retesting would
have resulted in lower HC and CO proposed standards (representing
the Clean Air Act's mandated 90 percent reduction). The inclusion
of the three additional vehicles would have further reduced the HC
standard.
The staff recommends retaining the standards (0.8 g/mile HC
and 10 g/mile CO) proposed in the LDT NPRM even though they are
higher than standards calculated from EPA's 21 vehicle sample. The
proposed standards represent an 88 percent reduction in HC emis-
sions and an 87 percent reduction in CO emissions as measured from
the final 21 vehicle baseline. Since these values are close to the
desired 90 percent reductions, the staff considers this action as a
reasonable option to follow. To repropose the standards at lower
levels would delay implementation of the rulemaking while making
only small environmental gains.
-------
Table 1
L.U.T. bAStLlNfc EMISSION RESULTS
02-07-80 09:27:05
1
2
u
<<
5
6
/
ft
9
10
11
12
1J
Ib
lo
17
DODGE
DODGE
DODGE
FOWL
rum,
F Oftl.'
FURU
Crttv
CnEV
CnFV
Cn£\<
CntV
GMC '
Gift
225
22b
318
3lh
302
3bO
160
360
360
360
360
2bO
307
3bO
3bU
3bO
350
404
41d
444
618
610
4?J
613
441
607
450
419
427
602
* LOT
2.20
1.20
3.30
3.30
7.10
24.20
24,20
24.20
24.20
24.20
24.20
9.40
28. '»0
7.80
7. HO
7.. SO
7.80
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
2
2
2
1
6
b
6
6
6
6
1
1
4
<*
4
s
X
-
s
s
=
=
=
s
It
1.1(10
1.100
1.6SO
1.050
7.100
4.033
4.UJ3
4.033
4.0.J3
4.033
4.033
9.400
2d.400
1.950
1.950
1.950
1.9bO
UTIAL &A!
CORK.
1
K32
U99
8.54
4,8b
4.tib
4.8b
11.31
34.18
2.35
2!3b
2.35
5EL1NE DATA
7.651
5.736
7.856
H.440
7.993
11.667
10.663
5.177
6.969
7.963
2.946
4.374
9.111
13.844
7.767
7.188
8.623
0.101
0.076
0.1S6
0.22H
0.683
0.566
O.blrt
0.2S1
0.3Jd
0.386
0.143
0.495
3.114
0.325
0.182
0.169
0.202
70.070
158.213
Bb.976
102.562
141.764
203. Ib7
2H.724
77.388
63.785
60.709
44.556
57.264
95.517
90.393
149.356
100.387
106.630
0.928 5.488
2.094 2.328
1.707
2.036
12.112
9.860
10.665
3.756
3.096
2.947
2.163
6.478
32.650
2.121
3.505
2.356
2.502
4.354
3.918
1.733
2.309
1.673
4.974
5.054
2.904
2.767
4.546
3.618
3.389
2.140
2.818
3.155
0.073
0.031
0.086
0.078
0.148
0.112
0.081
0.241
0.245
0.141
0.134
O.S14
1.236
0.080
0.050
0.066
0.074
18 JrtC
34b faUl
0.70 / 1
0.70U
0.84
14.742
0.124 155.920
1.313
1.958
0.016
83.100 100.00
8.058
102.29
3.408
PROGRAM NAME: SGwK57140C-LOT
-------
Table 2
19o9 L.O.T. HftStLlNE EMISSION KF.SULTS
02-07-60 09:38:22
1
2
3
4
5
6
7
8
10
11
12
13
It)
OODGt*
OUDOE
OOO&E
OOnCE
FORD
r OHi)
FOPO
FUPU
FOHU
rOPO
FGHD
CnEV
V
CMEV
CMFV
CHF.V
CnjiV
GMt
i
1HC
I
*
Ql!2
~225
225
31H
302
360
360
360
360
360
360
250
.307
350
3SO
350
3bO
345
* LOT
Ji£H2. SiLLS
42«
4<»4
61B*
421
425
491*
610*
H41*
607
419
<*SO*
602
001*
2.20
2.20
3.30
3.30
7.10
24.20
24. ?0
24.20
24.20
24.20
24.20
9.*0
26. 40
7. HO
/.<*<>
7!(JO
0.70
/ 2 =
/ 2 =
/ 2 =
/ 2 =
/ 1 =
/ e =
/ 6 =
/ 6 a
/ 6 =
/ 6 =
/ 6 =
/ 1 =
/ I =
/ 4 =
/ 4 =
/ 4 =
/ 1 =
1.100
1.100
1.650
1.650
7.100
4.d33
<».033
4.033
4.033
4.033
4.033
y.tOO
24 .'•00
1.950
1.9SO
1.950
1.950
0.700
WITH HETE
CORK.
>
1.32
1.32
1.99
1.99
tt.34
4,a5
4^85
4.d5
11.31
3^.1
-------
1 OOUGE
2 DODGE
3 DODGE
4 OOOGE
5 FORD
6
7
FORD
FUKLI
8 FORi)
9 FOHO
10 FOWL)
11 FORD
12 CHEV
1J CHKV
14 CHEV
Ib CHEV
16 CriEV
17 CMEV
Id CHEV
19 CriEV
20 GMC
21 IHC
TAbLE 3
1969 L.D.T. BASELINE EMISSION HESULTS
02-19-80 09(42117
CJJi
225
225
316
31H
302
360
360
360
360
360
360
2bO
2bO
292
307
307
350
350
3bO
3bO
345
JitUi .
404
418
444
618
421
425
473
491
610
613
623
617
607
621
419
427
4bO
602
601
* LOT
SALES
if. 20
2.20
3.30
J.JO
7.10
24.20
24.20
24.20
24.20
24. 20
24.20
9.40
1.60
28.40
26.40
7.80
7.80
7.80
7.80
0.70
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
2
2
2
1
6
o
6
b
6
t>
2
2
1
2
2
4
4
4
4
1
=
=
=
s
s
s
s
s
s
a
s
=
/
1.100
1.100
1.650
1.650
7.100
4.033
4.033
4.033
4.033
4.033
4.033
4.YOO
<*.700
1.600
14.200
14.200
1.950
1.950
1.950
1.950
0.700
kLL VfcHICl
COR*.
*
1.30
1.30
1.95
1.95
6.38
4.76
4.76
4.76
4.76
4.76
4.76
5.55
5.55
1.69
16.77
16.77
2.30
2.30
2.30
2.30
0.83
.ES
4.238
7,651
7.656
11.480
4.096
7.963
2.946
4.547
5.294
b.92b
5,942
3.910
7.516
10.345
9.111
5.128
7.767
6.789
7.628
8.623
7.612
•
iAL£=a!fi
0.055
0.099
0.153
0.224
0.343
0.379
0.140
0.217
0.252
0.330
0.283
0.217
0.417
0.195
1.527
0.860
0.179
0.156
0.176
0.199
0.1)63
86.072
70.070
85.976
102.562
69.613
60.709
44.556
65.038
77.663
116.007
67.867
27.438
62.402
83.903
95.537
56.641
149.356
70.528
144,841
106.630
106.287
SALEzMIC NQq-ttTP SAL£rMI2
1.118 6.380 0.083
0.910 5.488 0.071
1.675
1,998
5.835
2.691
2.122
3.097
3.698
5.619
3.232
1.523
3.463
1.585
16.017
9.496
3.439
1.624
3.335
2.455
0.695
4.354
3.918
3.267
2.904
2.767
4.862
5.780
5.382
5.139
4.89b
3.421
3.717
3.616
3.231
2.140
2.825
2.365
3.155
2.272
0.065
0.076
0.274
0.138
0.132
0.232
0.275
0.256
0.245
0.272
0.190
0.070
0.607
0.542
0.049
0.065
0.054
0.073
0.019
84.700 100,00
6.464
76.025
3.307
PROGRAM NAME: SGWK:7140C-LOT
-------
Table 4
Sales-Weighted Emission Results
Data Sample Size HC (g/mi) CO (g/mi) NOx (g/mi)
Initial Test 18 8.058 102.29 3.408
Results (be-
fore retesting,
Table 1
Initial Test 18 6.866 80.542 3.958
Results (after
retesting,
Table 2
Final Test 21 6.464 76.025 3.807
Results (in-
cludes all
vehicles tested,
Table 3)
-------
Table 5
90 Percent: Reduction from Baseline
HC and CO Levels
Data
Sample Size
18
(HC g/mi)
0.8*
CO (g/mi)
10*
Initial Test
Results (Table 1)
Initial Test 18 0.7 8
Results after
Retesting (Table 2)
Final Test Results 21 0.6 8
(Table 3)
* These levels were proposed as 1983 emission standards for
light-duty trucks in July, 1979 (FR, Vol. 44, No. 135).
-------
b. Comment s
Many of the issues concerning the LDT baseline program were
previously raised in connection with the 1984 heavy-duty engine
gaseous emissions rulemaking. These issues are analyzed in depth
in the Summary and Analysis of Comments document accompanying the
final heavy-duty rulemaking, and to the extent relevant, are
incorporated herein by reference. Because of this, the issues will
be treated breifly here, and the reader is referred to the heavy-
duty document for further discussion.
MVMA and the manufacturers criticized the light-duty truck
test program's sample size of 18 vehicles as being too small to
formulate an accurate baseline. The staff disagrees. EPA's
rationale for using 18 vehicles was based on two considerations.
First, the sales-weighted emissions after 18 vehicles had been
tested were leveling off at essentially constant values and EPA
believed that further testing would not significantly change the
baseline. Second, the emission 'results obtained from the 18
vehicles agreed favorably with the previous emission results
obtained from other EPA test programs for 1969 light-duty trucks.
This rationale is further discussed in the report "1969 Light-Duty
Truck Baseline Program and 1983 Emission Standards Development"
contained in the docket.
Tables 6 and 7 show that after testing only 8 vehicles, the
emission results began to stabilize. (NOTE: these are plots of
the 18 vehicles' emission results before retesting of 10 of the
vehicles).
The baseline results after retesting are:
6.87 g/mile HC
80.54 g/mile CO
Three additional vehicles were tested after issuance of the
NPRM. Inclusion of these three vehicles reduced the baseline by
less than 6 percent (see Table 3). Thus, final baseline emissions
for all 21 vehicles tested at EG & G are 6.464 g/mile HC and 76.025
g/mile CO. Inclusion of sales-weighted emission data from 11 LDTs
from other EPA programs (see Table 8) yields baseline values of
6.53 g/mile HC and 74.55 g/mile CO for the 32-vehicle sample. The
values for the 32-vehicle sample are 5 percent lower for HC and 7
percent lower for CO than the 18-vehicle sample.
The pertinence of constructing this 32-vehicle composite
baseline is to show that the 18-vehicle baseline results are
reasonably close to the 32-vehicle baseline results. Nearly
doubling the sample size had minimal effects on the baseline
results (i.e., 6.87 g/mile vs. 6.53 g/mile for HC).
The second consideration was the agreement between previous
1969 LDT emission test data and the baseline emissions data. The
-------
Table 6
LOT SflLES-WEIGHTED BRSELINE
EMISSIONS HC-(GRRMS/MI)
-------
Table 7
I
C.
o
CJ
3
R
§
—
o
o
o
LOT SflLES-HEIGHTED BflSELINE
EMISSIONS CO(GRRMS/MI)
-c—
4.CO
8.CO
lira
-------
Table 8
H69 L.O.T. riASELI'Nc
N Uc.SULTr> - COMPUSlTc! i)ATA
02-22-60 14:09:33
Vf H 1 .-1 I."
i r.u'-iijc
2 DOOGF
3 DODGF:
H CODGF.
•= FOPD
t FUPO
7 fUPO
H F0"0
^ FOO
il F'.'-^i
11 F : J r" 0
i .•: r" 0 -' "
13 i'J-'U
1-a r'Jr-0
16. F'JPO
; 7 ~ H '-' \i
U TnKV
1^ CrtKV
20 GM
': rn<-v
22 C"FV
2J f-f,
2* CMfc'V
2S Cr<£V
2o CHrlv
2'-> r-vc
,>'/ ,',v,C
30 r,."!C
31 GMC
J2 IMC
CiD '
225
225 .
318
3 lr,
2*0
302
360
J60
3 b'.1
360
3 60
36-J
JoO
360
36u
3 si;
25 n
•in.*
292
2v?.
30V
-j , j /
:>.../
3bO
3t>ii
3:>u
3bO
150
3-0
3^0
396
345
JFM"
404
428
4lft
4*4
8
018
421
•s,'-:5
-+73
4^1
blO
oi3
3
7
10
i*
4^1
t>2 .1
617
6
607
••>2i
12
*L9
427
Hr.il
602
1
2
11
9
601
, 1 M T
s HJ 1
2.20
2.?0
3.30
3.30
5.90
7.10
^4.20
^4.20
ir* .20
? 4 . ',1 'J
/.'• .2 u
2;>.2lJ
£<* .20
24.Pii
24.20
J.SO
•y.v.i
v.*j
1.60
1 .60
/-O .*'!
2-.: .'»!'
2b.nO
l.H(j
/.80
/.d'J
7.40
7.30
7.80
7.hO
0.80
0.70
/. 2 =
/ 2 =
/ 2 =
/ 2 =
/ 1 =
/ 1 =
/ 9 =
/ 9 =
/ •* •=
/ -y -
/ 9 =
/ 9 =
/ 9 =
/ 9 =
/ 9 =
/ I =
/ 2 =
/ 2 -
/ 2 =
/ 2 =
/ 3 -
.' 3 =
/ 3 =
/ 7 =
/ 7 =
/ 7 =
V —
/ 7 =
X 7 =
/ 7 =
/ 1 =
/ 1 =
1.100
1.100
1.6SO
l.o'^O
5.^00
7.100
4,.f,HS)
<^. t>89
2.t- 89
2.^i9
^ . fc 'j •>
2.oc9
£f . 6 J5 9
2.6H9
2.tWjy
3.500
4.700
4.700
0.800
O.hOO
9.4^7
9 . * "i 7
9.*b7
1.114
1.114
l< ;i4
1.114
1.11*
1.114
1.114
0.800
O./OO
C (\'rJ-/
^ltf\f< •
.*
1.16
l.lo
1.74
1.74
6.2J
7.48
2.rt3
2.83
2 . B 3
2 .^3
2.o3
2.«J
2.aJ
2.rt3
2.-»3
3.^,9
4.95
4.9b
O.b*
0.8*
9.96
9,^a
9.9B
1.17
1.17
) . 1 7
1.17
I. It
1.17
1.17
O.ti*
0.74
<•— r*
(ULibiriilti iALL-v.-TQ
4.23b
7.6^1
7.856
11 .480
6.8^0
4.096
7.963
2.9*6
*.•?* 7
S.?9*
0.9^5
5. s^«+2
4.530
8.0*0
12.4^0
t . 3 1 0
3.910
7 . b ) 6
10.345
6. >MO
9.111
5. 128
0.220
7.767
0.769
7 .62tt
tt . r> Z 3
4.9.40
9.000
9. /30
7.070
7.612
0.049
0.089
0.137
0.200
0.428
0.306
0.226
0 . ("• 8 3
0.129
0. ISO
0. 19o
0 . 1 68
0.1 2M
0.228
0.354
0. 159
0.194
0.372
0.08f
0.052
0.909
0.512
O.o<:0
0.091
0.080
0.090
0.101
0.058
0. 106
0.114
0.060
0.056
NON-iTQ 5
86.072
70.070
«5.976
102.562
114.970
69.613
60.709
4*. 556
6b.038
77.663
118.007
6/.H67
S6.000
103.500
106.460
54.480
27.43*
t>2.402
83.903
68.390
95.537
56.641
31.490
149.356
70.526
144. M*l
106. 63u
89.240
113.^00
Ib2.730
83.440
108.267
iM.t-«TO
0.99ft
0 . ft 1 2
1.49S
1.783
7.148
5.208
1.720
l.?62
1.&43
2.200
3.3*4
1.923
1.5H7
2.933
3.016
2.009
1.359
3.091
0.707
0.577
9.53"
5.650
3.141
1.754
0.8?H
1.701
1.252
1.048
1.334
1.793
0.703
0.799
• ii2lir iili 5 A i. K - w T o
6.380
5.*oB
4.354
3.91*
5.40o
3.267
2.904
2.7t>7
4.io2
5. 7rjO
5. Jo2
b. 1 34
4.1*0
2.H10
6.960
8.460
4.69-j
3.421
3.717
4.810
3 . n 1 8
3.^31
7.^HJ
2.1*0
2.i25
2.365
3.155
7.030
5.060
S.4SO
7.060
2.272
0.074
0.064
0.076
0.06ft
0.336
0.244
0.032
0.07H
0 . 1 3S
0. 16*
0.1S?
0.1*6
0.117
0.080
0.197
0.312
0.242
0. 169
0.031
0,041
G.jM
0. 3^^
0.722
0.025
0.033
0 .028
(i . 0 3 7
O.u33
0.. 060
0.06*
0.060
0.017
v*.900 100.00
6.S32
74.548
4.623
?: SGwK:7140C-L!)T
-------
Table 9
Estimated
No.
1
2
3 '
4 .
5 •
6
7
8
9
10
11
12
Manufacturer
General Motors
General Motors
Ford
Ford-
Dodge
General Motors
Ford
Ford
General Motors
Ford
General Motors
General Motors
1M9 Lieht Dutv Truck-Baseline .
Engine
350
350
360
390
383
292
360
240
396
360
350
307
tn3
in3
in3
in3
in3
in3
t»3
in3
in3
in3
in3
in3
•(6,000 - 8,500 GVWR)
Inertia Road
5000
•5000
• • 4500
. 5000
5000
5500
5000
' 4500
5000
5000
..5500
• 5000
Ibs' ' '
Ibs •
Ibs
Ibs
Ibs
Ibs
Ibs .
Ibs
Ibs '•'
Ibs
Ibs '.
Ibs
Average
17.9
17.9
13.1
17.9
17.9
22.7
17.9
21.8
21,1
21.1
22.7
17.9
•MMMMMV-V
19.2
Load
hp.
hp
hp
hp'"
hp
hP-- :
hp '
hp
hp
hp '
hp
>
•
Metric (g^kmj
Emissions g/mile
HC CO ' NOx
.'4.94
'9.00
4.53
4.31
8.54
6.18
8.04
6.89
7.07
12.49
9.73
6.22
.'7.33
4.55
• 89.24
• 113,
56.
54.
149.
\ 68.
103.
114.
83.
106.
• 152..
31.
• 93.
58.
60
00
48.
00
39
50.
97
44
46
73
4.9
61
17
7.03
5.08
4.14
8.46
9.12
4.81
2". 81
5.40
. 7.06
6.96-
•5.49
7.24
6.13 '
3.81
Sources: A Study of- Baseline Emissions on 6,000-14,000 Pound Gross Vehicle Weight Trucks,
June 1973, Automotive Environmental Systems, Inc., APTD-1572, (Vehicles 1 to 5)
Baseline Emissions on 6,000 to 14,000 Pounds Gross Vehicle Weight Trucks, June, 1973,
Southwest Research Institute, APTD-1571 (Vehicles 6 and 7)
Medium Duty Baseline Tests, Environmental Protection Agency, Unpublished (Vehicles 8 to 12)
-------
sales-weighted baseline results for the initial tests of 18 ve-
hicles (Table 1) are 8.06 g/mile HC and 102.29 g/mile CO. This is
close to the average emissions (7.33 g/mile HC, 93.61 g/CO) for 12
vehicles tested previously in EPA programs (Table 9). The revised
baseline results agree even more closely than did the original
baseline data.
Commenters criticized the vehicle sampling plan for inadequate
stratification. They cite that only 1 GM 307 (28.4 percent of LDT
sales) was tested while 4 GM 350's (7.8 percent of LDT sales) were
tested. EPA originally, because of the sales data available, set a
target number of GM 350's to be tested at 2-3. Later this was
revised to two GM 350's when more accurate sales data was provided
by the manufacturers. Since 4 vehicles were already procured, they
were included.
Regarding the GM 307, EPA's original target goal was seven
engines, however only two were finally obtained. This was due to
our inability to locate more engines. Only one of these engines
was tested at the time the original baseline testing was completed
for the NPRM. The second 307 was tested subsequently, and that
data is included in Table 3.
In evaluating the adequacy of the samples tested for in-
dividual families it is important to bear in mind the sales-
weighting process used by EPA to determine the baseline emission
rate. In that process, the emission results from each engine
are weighted to account for the fraction of sales which the
subject engine family represents. Because of this, it is not
in any way necessary for the number of engines tested in each
family to also be sales-weighted. A baseline could just as
well be computed from a sample containing equal numbers of engines
from all families. The reason why EPA desired to sales-weight the
sample of engines procured was to obtain the most accurate data for
the most important (in terms of sales) families with the minimum
overall number of engines. Failure to be completely successful at
this goal does not invalidate the baseline process.
MVMA contends that the sample vehicles are unrepresentative
since: 1) they were all procured from the San Antonio, Texas area;
2) the mileages were much less than the new proposed useful life;
and 3) the 18-vehicle sample represented ony 84 percent of LDT
sales.
The staff disagrees with this assertion. The vehicles
selected for baseline testing were 1969 model year light-duty
trucks which had received no major engine overhaul and which had
original carburetors and distributors in most cases. In addition,
the vehicle selected had to pass selection criteria aimed at
obtaining vehicles in good mechanical condition. Pre-test main-
tenance assured that the vehicles were representative. Speci-
fically, engines were tuned to manufacturers specifications in an
effort to preclude uncharacteristic emissions caused by out of
13;
-------
spec, engines. Defective components were replaced in some in-
stances. Accordingly the pre-test maintenance normalized the
vehicle sample. MVMA has presented no reasons why emissions from
well maintained vehicles would have significant geographic vari-
ability.
Commenters were critical of the test vehicle mileages because
they were less than the proposed useful life definition. EPA's
goal in formulating a baseline has been thoroughly discussed in the
Analysis of Comments to the heavy-duty 1983 gaseous emissions
rulemaking (issue M). In summary, all efforts were made to insure
that only mechanically sound engines with original equipment were
used in the baseline to accurately reflect 1969 engines. Practical
limitations then resulted in low-mileage applications being prefer-
entially selected. In-use deterioration of we11-maintained vehi-
cles is inherently low (see discussion below), so that no important
i mileage related effect will occur.
MVMA criticized the 18 vehicle sample even though it repre-
sented nearly 84 percent of light-duty truck sales. EPA believes
that 84 percent is a reasonable figure for determining a baseline.
The emission results for 21 vehicles tested at EG&G are 6.464
g/mile HC and 76.025 g/mile CO and represents 84.7 percent of the
LOT sales (6001-8500 Ib. GVWR). EPA's 32 vehicle sample results in
emissions of 6.532 g/mile he and 74.548 g/mile CO and represents
94.9 percent of the LDT sales. The difference in emissions between
the two baselines is minimal.
Commenters contend that the pre-test vehicle maintenance was
improper for establishing the baseline. Maintenance should have
been limited to only maintenance specified in the certification
procedure as defined in the new NPRM. The staff disagrees with
this contention.
The maintenance performed on the test vehicles was necessary
to assure that the test engines were as close to new engine con-
figuration as possible without major overhaul. If a component was
so worn that manufacturers specifications could not be met, then it
was rebuilt or replaced. However, the most important consideration
was to tune-up the engines to 1969 manufacturers specifications.
In fact, several vehicles were rejected after having been procured
because major engine maintenance was needed. The maintenance
performed was not limited to certification specified maintenance.
the staff believes the maintenance performed was reasonable,
however, for 10 year old vehicles.
Manufacturers maintained that deterioration factors should
have been calculated and applied to the baseline and standards.
The staff disagrees. Baseline emissions were calculated to
represent the average of the actually measured emissions from
1969 model year light-duty trucks. Since the engines were charac-
terized as new engines (due to pre-test maintenance) a zero deteri-
oration factor assumption was made. Furthermore, certification
-------
data for non-catalyst engines shows that these engines have
inherently low deterioration factors, and support the zero d.f.
assumption.
MVMA and the manufacturers asserted that sampling error should
be recognized in calculating the standards and recommended using
the upper 90 percent confidence limit of the mean. The staff does
not believe that this procedure is required by the Clean Air Act.
According to section 202 a(3)(A)(ii) of the Act, the standards are
to be derived "from the average of the actually measured emis-
sions. "
MVMA contends that as a result of other elements of the NPRM,
the standards are more stringent than a 90 percent straight reduc-
tion. The staff disagrees with this comment. EPA is directed to
promulgate regulations which will cause at least a 90 percent
reduction in HC and CO emissions as measured from 1969 model year
LDT's. The other elements of the rulemaking do not directly affect
the stringency of the numerical standards, which are unchanged.
Rather, they work toward a fuller degree of compliance with the
standards. Thus, they help assure that the desired minimum per-
centage reductions are attained.
4. Recommendation
Retain the new emission standards as proposed.
-------
L. Issue: Fuel Economy
1. Summary of the Issue
If adopted as proposed, will the regulations and standards
cause complying vehicles to experience a fuel economy penalty?
2. Summary of the Comments
In general, most commenters preferred to analyze fuel economy
effects on the basis of 40 percent AQL and a 50,000-mile useful
life, in spite of the proposed changes. Bearing this in mind,
most manufacturers considered the current emission standards in
California to be approximately equivalent in stringency to the
proposed standards; the impact on future fuel economy was extra-
polated from the fuel economy performance of the California fleet.
No fuel economy impact on diesel vehicles was claimed.
r
General Motors claimed a one mpg (7 percent) fuel economy
penalty in gasoline vehicles as a direct result of these standards.
The addition of air injection and danger to spark retard, EGR, and
carburetor calibrations to assist catalyst lightoff are all
strategies currently used in California which GM argued would
be adopted Federally to meet the proposed standard. No closed-loop
systems are anticipated.
Ford claimed that use of oxidation catalyst, EGR and air
injection would be necessary to meet the standards at a 40 percent
AQL and 50,000-mile useful life. In addition, the higher inertia
powertrains would require light-off catalysts of dubious dura-
bility. Overall fuel economy penalties of 3.4 percent and 7.2
percent for models with and without light-off catalysts respective-
ly were anticipated. No projections were given under 10 percent
AQL, full useful life conditions.
International Harvester claimed that EGR, additional air
injection, and oxidation catalysts would be needed.
Volkswagen claimed that with a 40 percent AQL and 50,000-mile
life, only EGR and oxidation catalysts would be necessary. (The
lean burn techniques presently used permit the reduction of NOx in
the exhaust with only oxidation catalysts.) No additional air
injection would be necessary. With 10 percent AQL, however,
Volkswagen's preferred strategy would be the use of three-way
catalyst and feedback carburetor. In general, Volkswagen foresaw a
5 percent fuel economy penalty in its heavier vehicles, as opposed
to some gains using three-way systems with lighter inertia weight
vehicles.
American Motors predicted a fuel economy loss of .4 to .8 mpg
(40 percent AQL, 50,000-miles). No hardware changes were pre-
dicted; changes to carburetor calibrations would allow compliance
but result in the fuel economy penalty.
-------
In public testimony, commenters disclosed under questioning
that electronic engine control technology was being pursued, but
refused to give further details for "proprietary" reasons.
3. Analysis of the Comments
Since most commenters preferred to analyze this issue on the
basis of the California fleet, the staff's initial analysis of
expected fuel economy impact will follow suit. From there, an
analysis will be performed of the differences between California
compliance strategies and those strategies which the staff antici-
pates to be used to comply with the 1983 standards. An estimate of
the fleetwide fuel economy impact of these proposed regulations
will conclude the analysis.
Table L-l presents comparative emission and fuel economy data
for California and Federal light-duty trucks tested at EPA's Motor
Vehicle Laboratory. In all cases (except where noted), the compar-
ative vehicles are identical with the exception of emission control
equipment.* For ease of comparison, the control strategies adopted
for California compliance and their impact on the California
vehicle's fuel economy (or averaged over all similar-technology
tests) are presented in Table L-2.
Tables L-3 and L-4 present data used in deriving the staff's
projected fuel economy penalty for the national light-duty truck
fleet presuming that no new technology is introduced in 1983. The
analysis sales-weights the penalties derived from EPA certifiation
data in Table L-l with one exception,** and derives a fleet-wide
fuel economy penalty based upon the projected market shares of
four, six, and eight cylinder engines in 1983.
The results of the comparison support the industry's conten-
tion that an overall fuel economy penalty is likely if California-
style technology is applied. The staff's analysis indicates that
with no technological improvements, an average penalty of 5.2
percent per manufacturer was experienced in California. This
agrees fairly well with the commenters' projections. The staff
takes strong issue, however, with the claim that such a penalty is
unavoidable for compliance with the 1983 standards.
* Same inertia weight, transmission, axle ratio, and dynamometer
road load horsepower.
** Some eight cylinder engines tested at EPA's lab indicated no
fuel economy penalty. However, vehicles in higher weight classes
are certified to California standards which are roughly equivalent
to the Federal standards, i.e., no reduction, no applied tech-
nology, no fuel economy loss. Based upon GM's comments alluding to
the fact that air pumps would be needed on eight cylinder families,
an overall penalty of 4 percent was attributed to the air pumps'
parasitic horsepower requirements.
-------
Table L-l
Emission and FOB! Economy Data for I.ighc-Duty Trucks
Acquired From EPA's Emission Certification Tests for 1980 MY
Vehicle
Model
Jeep CJS
Jeep CJ7
Jeep CJ7
Jeep CJ7
Jeep CJS
Jeep CJS
Eagle Wagon
Eagle Wagon
Jeep CJ7
Jeep CJ7
Jeep CJS
Jeep CJS
Jeep J20
Jeep J20
Cherokee
Uagoneer
Bl Van
(109" HB)
B100 Van
(109" WB)
B2 Van
(127" HB)
B2 Van
(127" WB)
B2 Van
(127" WB)
W150
HI 50
F-100 Reg.
CAB LWB
F-100 Reg.
CAB LWB
F-ISO SC LWB
F-150 SC LWB
F-250, 4x4 LWB
F-250, 1x4 LW?
F-150, 4x2 LWB
F-150, 4x2 SWB
F-150, 4x2 LWB
F-150, 4x2 SWB
E-250, LHB
Cluhuagon
E-250, LWB
Clubwagon
F-100, 4x2
SWB RC
F-100, 4x2
' SWB "RC
F-100, 4x2
1MB RC
F-100, 4x2
LWB RC
ENGINE
MFC
AMC
AMC
AMC
AMC
AMC
AMC
AMC
AMC
AMC
AMC
AMC
AMC
AMC
AMC
AMC
AMC
Chrysler
Chrysler
Chrysler
Chrysler
Chrysler
Chrysler
Chrysler
Ford
Ford
Ford
Ford
Ford
Ford
Ford
Ford
Ford
Ford
Ford
Ford
Ford
Ford
Ford
Ford
Family
BT6C1
BT9A1
CT4W1
CT3A1
CT4W1
CT3H1
CT4W1
CT3A1
HT3V1
HT3A1
HT3V1
HT3A1
NT3A1
NT3A1
NT3A1
HT3A1
OTA-225
-1-BXP
OTA-225
-1-BCP
OTA-3 18/360
4 BCP
OTA-3 18/360
4 BFP
OTA-3 18/360
4 BFP
OTA-318/360
4 BCP
OTA-318/360
4 BFP
4.9ND
4.9HA
4.9ND
4.9NA
5. OMB
5.0NA
5.0NB
S.ONA
5.0NG
S.ONA
5.0NB
5. ON A
S.OtlC
S.ONA
5.0NG
S.ONA
CID
151
151
258
258
258
258
258
258
304
301
304
304
360
360
360
360
225
225
360
360
360
360
360
300
300
300
300
302
302
302
302
302
302
302
302
302
302
302
302
Fed. or
Calif.
Both
Fed.
Calif.
Fed.
Calif.
Fed.
Calif.
Fed.
Calif.
Fed.
Calif.
Fed.
Calif.
Fed. ,
Calif.
Fed.
Calif.
Fed.
Calif.
Fed.
Fed.
Calif.
Fed.
Calif.
Fed.
Calif
Fed.
Both
Fed.
CjUf.
Fed.
Calif.
Fed.
Calif.
Fed.
Calif.
Fed.
Calif.
Fed.
Additional
Technology in
California
3-way Cacalysc
Light-off
Catalyse
Light-off
Catalyse
Light-off
Catalyse
Increased Cata-
lyst Loading
Increased Cata-
lyst Loading
Calibration
Calibration,
Axle ratio
Catalyst Loading,
Inereia Weight,
Axle Ratio
Light-off Cata-
lyst (Fed.), Cata-
lyst Loading, Iner-
tia Weight, Axle
Ratio
Transmission, Iner-
tia, Axle Racio,
Light-off Catalyst,
Catalyst Loading
Light-off Catalyst,
Axle Ratio
Light-off Catalyst,
Axle Ratio
Larger Catalysts,
Transmission,
Inertia Weight
Larger Catalyst,
Inertia Height
Light-off Catalyst,
Larger Catalyst
Larger Catalyst,
Calibration
Light-off Catalyst,
Larger Catalyst,
Axle Ratio
Manual Transmis-
sion., Axle Ratio,
Lifhc-off Catalyst,
Larger Catalyst
4X
Cert i f icat ion
HC
0.18
0.42
0.43
0.48
0.40
0.67
0.28
0.35
0.38
0.94
0.28
0.73
0.57
0.56
0.56
0.72
0.23
0.58
0.22
0.40
0.46
0.37
u.78
0.23
0.57
0.24
0.87
0.59
0.73
0.42
0.53
0.24
0.58
0.40
0.51
0.22
0.61
0.24
0.59
CO
2.47
5.62
7.39
11.28
4.43
6.35
3.88
7.76
3.90
9.27
2.80
4.87
9.10
11.00
9.90
15.00
1.81
11.00
6.99
8.80
8.10
8.80
14.14
2.10
i.OO
6.20
8.80
4.82
6.92
3.96
9.62
2.10
9.62
6.72
4.13
1.60
7.60
2.00
3.46
Levels
NOx
1.30
1.50
1.00
1.40
1.41
2.00
1.50
2.00
0.92
1.60
1.45
1.70
1.74
1.65
1.94
1.36
1.28
1.60
1.20
1.20
0.94
1.70
1.40
1.90
1.08
1.60
1.86
1.59
1.59
1.42
1.31
1.45
1.31
1.17
1.50
1.45
1.97
1.45
1.78
MFC
21.5
21.4
15.1
15.6
15.4
15.2
15.0
16.4
12.1
14.2
11.4
12.9
10.7
11.0
11.5
11.8
17.1
15.4
11.6
11.4
13.1
11.8
10.8
16.8
16.2
16.0
17.9
12.4
12.8
14.4
14.7
13.5
14.7
11.8
12.5
13.2
16.4
13.0
15.4
Percent
Oi f ference
+0.5
-3.2
*1.3
-8.5
-14.8
-11.6
-2.7
-2.5
+11.0
+1.8
-11.5
»9.3
+3.7
-10.6
-3.1
-2.0
-8.2
-5.6
-19.5
-15.6
13 (
-------
Table L-l (cont'd)
Emission and Fuel Economy Data for Light-Duty Trucks
Acquired From EPA's Emission Certification Tests for 1930 MY
Vehicle
Model
-
Traveler 4x4
Traveler 4x4
Scout II, 4x4
Scout II, 4x4
Scout II, 4x4
Scout II, 4x4
Chevy UJV-1
Chevy LUV-1
Chevy LUV-4
Chevy LUV-4
Deluxe, 5 spd
Pickup
Deluxe, 5 spd
Pickup
Pickup, 5 spd
King Cab
Pickup, 5 spd
King Cab
Pickup King Cab
Pickup King Cab
Pickup Cab
and Chassis
Pickup Cab
and Chassis
D50
D50
DSO
DSO
D50
050
D50
DSO
Courier
Courier
Courier
Courier
Courier
Courier
Land Cruiser
Hardtop
Land Cruiser
Hardtop
Land Cruser
Sta. Sgn.
Land Cruiser
S/W
4-WD Pickup SB
4-UD Pickup 13
ENGINE
HFC
CM
CM
IHC
IHC
me
IHC
IHC
IHC
Isuzu
Isuzu
Isuzu
Isucu
Nissan
Nissan
Nissan
Nissan
Nissan
Nissan
Nissan
Nissan
Mitsubishi
Mitsubishi
Mitsubishi
Mitsubishi
Mitsubishi
Mitsubishi
Mitsubishi
Mitsubishi
Toyo Kogyo
Toyo Kogyo
Toyo Kogyo
Toyo Kogyo
Toyo Kogyo
Toyo Kogyo
Toyota
Toyota
Toyota
Toyota
Toyota
Toyo t a
Family
08K4AA
08K4G
V304
V304
V345
V345
4-196
4-196
AITC
AITB
AITC
AITB
TL20C
TL20F
TL20C
TL20F
TL20C
T120F
TL20C
TL20F
G5T-C
G52T-F
G5T-C
G52T-F
C5T-C
G54T-F
G5T-C
G52T-F
CHAT
CHAT
CWBT
OWBT
CWBT
CHBT
2F(C>
2F(F>
2F(C)
2F(F)
JOR(TC)
20K(TF)
CID
400
400
304
304
345
345
196
196
111
111
111
111
119
119
119
119
119
119
119
119
121
121
156
156
156
156
156
156
120
120
140
140
140
140
258
258
258
258
134
134
Fed. or
Calif.
Calif.
Fed.
Calif.
Fed.
Calif.
Fed.
Calif.
Fed.
Calif.
Fed.
Calif.
Fed.
Calif.
Fed.
Calif.
Fed.
Calif.
Fed.
Calif.
Fed.
Calif.
Fed.
Calif.
Fed.
Calif.
Fed.
Calif.
Fed.
Calif.
Fed.
Calif.
Fed.
Calif.
Fed.
Calif.
Fed.
Calif.
Fed.
Calif.
Fed.
Additional
Technology in
California
Lower Catalyst
loading, Air Pump
Calibration,
Manual Transmission
Manual Transmis-
sion, More Inertia
Woight High Axle
Ratio, Calibration
Extra Speed Trans-
mission, Axle
Ratio, Calibration
Oxidation Catalyst,
Calibration
Oxidation Catalyst,
Greater Air,
Higher Catalyst
Volume and Loading
Greater Air,
Higher Catalyst
Volume and Loading
Greater Air,
Higher Catalyst
Volume and Loading
Greater Air,
Higher Catalyst
Volume and Loading
Pulse Air
Calibration
Calibration
Calibration
Extra Speed Trans-
mission, Calibration
Calibration
Calibration
Calibration
Calibration
Oxidation
Catalyst
4K
Certification
IIC
0.41
0.54
0.65
0.61
0.78
0.57
0.55
0.50
0.22
1.40
0.23
1.40
0.26
0.73
0.24
0.83
0.18
0.51
0.14
0.62
0.26
0.57
0.19
0.75
0.26
0.17
0.20
0.29
0.24
0.40
0.24
0.34
0.23
0.26
0.22
0.44
0.24
0.49
0.15
1.19
CO
5.18
12.74
4.98
4.72
6.80
5.52
8.90
6.80
4.29
14.00
2.73
12.00
2.97
9.00
2.16
12.00
2.61
6.50
2.61
9.50
3.83
7.64
2.49
7.17
3.54
3.42
3.16
4.20
2.37
6.23
2.50
7.56
2.57
6.72
3.40
8.73
3.10
9.49
1.89
13.82
Levels
NOx
1.50
1.30
1.36
1.75
a. oo
1.50
1.70
2.00
0.98
1.60
1.30
2.00
1.00
1.64
1.00
1.35
1.10
1.74
1.40
1.54
1.20
1.70
1.10
1.80
1.10
1.20
1.20
1.70
1.10
1.49
1.20
1.56
1.20
1.56
1.18
1
1.80
1.42
1.70
1.40
2.00
MPG
11.8
11.8
10.8
11.5
12.0
12.3
14.9
15.8
21.9
23.3
21.7
21.7
21.9
24. S
22.8
25.1
21.6
24.1
15.1
14.8
22.0
22.1
21.5
22.5
21.7
21.7
21.4
21.6
25.5
17.0
20.0
21.8
19.5
20.4
11.8
12.3
11.5
11.6
18.0
17.6
Percent
Difference
._
-6.1
-2.4
-5.7
-6.0
__
-10.6
-9.2
-10.4
+2.1
-0.5
-4.4
__
-0.9
-5.6
-8.3
-4.4
-4.1
-0.9
»2.3
-------
Table L-l (cont"d)
Eii in ion and Fuel F.conony Data for Light-Duty Trucks
Acquired From LPA's Emission Certification Tests for 1980 MY
Vehicle
Model
Long Bed
Pickup
Long Bed
Short Bed
Pickup
Short Bed
Pickup
Bus
Bus
Camper
Camper
Truck
Truck
Truck
Truck
Truck
Truck
ENGINE
HFC
Toyota
Toyota
Toyota
Toyota
Vtf
VW
vw
vw
vw
vw
vw
vw
vw
vw
Family
20R(TC)
20R(TF)
20R(TC)
20R(TF)
11
11
12
12
37PC
37PF
37PC
37PF
37PC
37PF
CID
134
134
134
134
120
120
120
120
97
97
97
97
97
97
Fed . or
Cnlif.
Calif.
Fed.
Calif.
Fed.
Calif.
Fed.
Calif.
Fed.
Calif.
Fed. t
Calif.
Fed.
Calif.
Fed.
Atldit ional
Technology in
California
Oxidation
Catalyst
Manual Transmission,
Oxidation Catalyst,
Calibrat ion
3-Way Catalyst,
Closed Loop Carbu-
retor, Ho EGR
3-Way Catalyst,
Closed Loop Carbu-
retor, Ho EGR
Oxidation
Catalyst
Oxidation
Catalyst
Oxidation
Catalyst
4K
Certification Levels
HC
0.16
0.96
0.20
0.64
0.19
0.91
0.28
0.79
0.16
1.30
0.31
1.20
0.14
1.40
CO
1.89
12.83
2.19
9.87
3.80
13.00
7.60
10.00
2.00«
5.73
1.10
6.06
1.30
6.47
NOx
1.10
1.70
1.00
1.60
0.70
2.00
1.10
1.50
*
1.40
1.50
1.10
1.80
1.10
1.80
MFC
18.6
19.9
18.0
22.2
16.5
17.2
15.7
16.8
22.0
21.6
22.4
22.8
23.2
22.8
Percent
Difference
-6.5
-18.9
-4.1
-6.5
+1.9
' -1.8
+1.8
-------
First of all, the technology applied in California was the
"add-on" type and incorporated no technological improvements. The
California strategies were designed to minimize production and
tooling changes, i.e., manufacturers needed to change the large
volume forty-nine state product to a "customized" product for the
single state with stricter emission standards and not unreasonably
chose the easiest method. Fuel economy was not as critical a
design parameter as the economies of scale associated with the
production of similar vehicles.
Secondly, as presented in Table L-6, the fleet-wide emission
reductions required for California were significantly greater for
HC and CO than those anticipated for 1983. Less than half of the
California HC reduction and 12 percent less of a CO reduction are
required. Only 5 percent more of a NOx reduction will be needed,
and it can be reasonably argued that the 5 percent is not so much a
reduction in emission levels but a reduction in emission vari-
ability.* Given the fact that less reductions are necessitated by
the 1983 standards, it follows that the fuel economy impact will be
similarly reduced. Experience tells us that HC control follows CO
control (i.e. CO is the more difficult of the two to control in a
catalyst system), hence a 12 percent less CO reduction should, at
the least reduce the fuel economy loss by a similar amount, i.e.,
from -4.51 to -3.97 percent per manufacturer.
The staff's analysis indicates that in the worst case, i.e.,
where manufacturers ignore fuel economy considerations in engine
design, and introduce no new technologies of the type already in
production on 1981 passenger cars utilizing essentially the same
power plants, then overall engine fuel economy will be reduced
approximately 4 percent per manufacturer.
It is the staff's belief, however, that this worst case
scenario is unlikely. New emission control technologies (in
particular, electronic engine controls) are being marketed today
which not only control emissions but also enhance fuel economy.
* No change in the statutory NOx standard has occurred, however,
changes in the AQL level from 40 percent to 10 percent imply that
either lower low mileage emission targets or lower variability will
be required. The staff's projected 1983 low mileage NOx target
of 1.4 g/rai is conservative and assumes high NOx variability on a
test-to-test basis during SEA audits. This assumption is judged
extremely conservative on the basis of audit data from the state of
California)._!_/ This data indicates that in 534 tests of GM, Ford,
IHC, and Chrysler LDTs, after application of EPA's estimated full
life deterioration factors, 9.4 percent of the vehicles exceeded a
NOx level of 2.35 g/mi; if a particularly variable Ford engine
family is discounted (5.8M/6,6NA), then in 441 tests only 4.3
percent exceeded the 2.35 g/mi limit. In short, the industry is
already meeting a 10 percent AQL level for NOx, no further reduc-
tion in design targets would be necessary, and therefore no fuel
economy penalty associated with NOx control is expected.
-------
Electronic engine controls are seeing, widespread use in
passenger cars in 1980, and in conjunction with three-way cata-
lysts, will see even more use in 1981 as tighter NOx and fuel
economy standards take effect. Two systems used today are General
Motors' C-4 (Computer Controlled Catalystic Converter) system and
Ford's EEC III system. Both systems control air/fuel ratio (for
use with a three-way catalyst), ignition spark timing, air injec-
tion, EGR flow, and evaporative canister purge. In addition, the
C-4 system controls idle speed and operation of the torque conver-
ter clutch; the EEC III also controls throttle kicker position.
Each system allows highly accurate optimization of the combustion
process over a wide variety of engine operating modes and condi-
tions.
Matrix mappings of emission generation and fuel consumption
are made for every possible combination of engine operating condi-
tions .J/4/ Those conditions which 'are either determined by the
driver ("for design purposes, the "driver" is the Federal Certifi-
cation Test Cycle), or which are environmental (engine temperature)
are then optimized for fuel economy and emissions by generating the
optimum combination of remaining engine parameters (e.g., spark
advance, EGR rate, fuel/air ratio, etc.). Over the full range of
operating conditions, these optimum parameters take the form of
continuous mathematical functions, which are then programmed
algorithmically into the control unit microprocessor.^/ The result
is an engine whose calibration continuously varies in response to
driver and environmental demands, and which is far superior to the
static mechanical calibrations now present on all light-duty
trucks.
It can be argued that engine calibrations significantly affect
fuel economy and emissions, as illustrated in Table L-2. Several
Federal engine families were merely recalibrated to meet the
California standards, i.e., no hardware differences exist. Engine
recalibration can lower the combustion temperature to reduce NOx
(spark retard), and increase the rejection of thermal energy to
speed catalyst light-off to reduce HC and CO; both effects reduce
combustion efficiency, i.e., degrade fuel economy. Electronic
controls can directly alleviate fuel economy penalties while also
controlling emissions. For example, continuously variable spark
timing allows cold start timing such that catalyst light-off is
enhanced, and to a much greater degree than a single static timing
position would allow. Once catalyst light-off has occurred (as
determined by a temperature sensor), the control unit switches the
timing to a more fuel efficient range. In short, both HC and CO
cold start emissions are controlled and overall fuel economy is
enhanced.
The magnitude of the fuel economy effect is dependent pri-
marily upon engine size, vehicle inertia weight, and applied
emission control technology. The National Highway Traffic Safety
Administration (NHTSA) in its preliminary analysis of proposed 1983
fuel economy standards for light-duty trucks2/ concluded that even
-------
Table L-2
Fuel Economy Effects of Several Emission
Control Strategies in California Light-Duty Trucks**
Number of Test
Data Points
on the Average
Additional
California
Technology*
3-way catalyst
with feedback
carburetors
Calibration only 8
Greater oxidation 8
catalyst volume,
loading, and
additional AIR
Light-off catalysts 3
alone
Light-off catalysts 9
with all other
changes
Overall average 38
without light-off
catalysts
Vehicle changes 5
only
Vehicle changes 15
with all other
changes
Overall average 47
Average
Fuel Economy Impact (%)
..Calif, mpg - Fed, mpg
Fed, mpg
-3.4
-3.2
-7.6
-3.5
-5.7
-3.9
-4.5
-4.4
-4.3
Standard
Deviation
of the Average
3.6
2.7
5.8
4.9
9.3
5.5
4.5
9.07
6.3
* NOTE: Where no differences in emission control hardware are
indicated between California and Federal vehicles (as evaluated
from EPA Certification records), and significant emission level
differences exist, this analysis assumes that a change in calibra-
tion (spark timing, distributor curve, carburetor setting) has
occurred. It is understood that calibration changes can and may
have occurred during the application of any other technology.
** This table makes no correction for 8-cylinder engine/higher
inertia weight vehicles, (see Table L-3).
-------
Table L-3
Projected Fuel Economy Impact of 1983 Regulations
(Using Current Engine Mix and No Technology Improvement)
Comparative
(Fed. /Calif.)
Families
(# cylinders)
Fuel
Economy*
Impact (%)
Projected
Sales
Fraction**
Sales
Weighted
F.E. Impact
AMC BT9A1/BT6C1 (4)
AMC CT3A1/CT4W1 (6)
AMC CT3H1/CT4W1 (6)
AMC HT3A1/HT2V1 (8)***
AMC NT3A1/NT3A1 (8)***
Chrysler 225 BCP/BXP (6)
Chrysler 318/360 BFP/BCP (8)***
Ford 4.9 NA/ND (6)
Ford 5.0 NA/NB,NG (8)***
GM 08K4G/08K4AA (8)***
Nissan TL20F/TL20C (4)
Mitsubishi G527F/G57C (4)
Toyo Kogyo OMAT/OMAT (4)
Toyo Kogyo OWBT/OWBT (4)
Toyota 2F(F)/2F(C) (6)
Toyota 20R TF/TC (4)
VW 37PF/37PC (4)
IHC V304/V304 (8)***
IHC V345/V345 (8)***
IHC 4-196/4-196 (4)
Isuzu AITAB/AITC (4)
+ .5
-3.2,-8.5
+ 3.2
-14.8,-11.6
-4.0,-4.0
+ 11
-4.0,-4.0,-11.5
+3.«7,-10.6
-3.1,-2.0,-5.6
-8.2,-19.5
-15.6
-4.0
+2.1,-10.4,-9.2
-10.6
-.5,-4.4,0,-.9
-5.6
-8.3,-4.4
-.9,-4.1
+2.3,-6.9,-18.9
+1.9,-1.8,+1.8
-6.1
-4.0
-5.7
-6.0,0
Totals =
Adjusted totals for sales not represented =
.0093
.0241
.0142
.0060
.0252
.0257
.0391
.0787
.1341
.2277
.0249
.0099
.0105
.0081
.0026
.0197
.0026
.0047
.0053
.0016
.0178
.6918
1.0000
+.0046
-.1410
+.0184
-.0792
-.1008
+.2825
-.2542
-.5430
-1.2717
-.9108
-.1749
-.0144
-.0589
-.0529
-.0066
-.1520
+.0016
-.0287
-.0112
-.0091
-.0534
-3.57
-5.15%
* From Table L-l.
** EPA certification records - manufacturer's projected sales for
1980.
*** It is assumed here that some 8-cylinder engines are used in
vehicles whose inertia weight is high enough to result in certifi-
cation to California standards comparably stringent to 1980 Federal
standards (i.e., fuel economy comparisons are not indicative of the
effects of the 1983 Federal standards). To compensate, a fuel
economy penalty of 4 percent is assumed for all vehicles exhibiting
less of a penalty. This 4 percent is based upon comments attrib-
uting a 4 percent loss on 8-cylinder engines due to air pump
addition.
-------
Table L-4
Impact of 1983 Regulations,
Using Projected Fleet Engine Mix
(Assuming No Technology Improvements)
Adjusted Sales Fractions*
4 cylinder
6 cylinder
8 cylinder
Totals
1980***
.151
.210
.639
.999*
1983
.15
.40
.45
1.00
Adjusted Fuel Economy Penalty
1980***
-.736
-.563
-3.85
-5.15
1983
-.731
-1.072
-2.711
-4.51
* Some totals do not add up to 1.000 due to round-off error.
** For 1983 sales projections, see Chapter 5 of the Regulatory
Analysis of this rulemaking.
*** From Table L-3.
-------
Table L-5
Sales Weighted (by Current Federal Sales) Emission
Contributions of 1980 Federal and California LOT Fleets
Comparative
Families (Fed./Calif.)
AMC BT9A1/BT6C1
AMC CT3A1/CT4W1
AMC CT3H1/CT4W1
AMC HT3A1/HT3V1
AMC NT3A1/NT3A1
Chrysler 225 BCP/BXP
Chrysler 318/360 BFP/BCP
Ford 4.9 NA/ND
Ford 5.0 NA/NB,NG
GM 08K4G/08K4AA
Nissan TL20F/TL20C
Mitsubishi G52TF/G5TC
Toyo Kogyo OMAT/OMAT
Toyo Kogyo OWBT/OWBT
Toyota 2F(F)/2F(C)
Toyota 20R TF/TC
VW 37PF/37PC
IHC V304/V304
IHC V345/V345
IHC 4-196/4-196
Isuzu AITB/AITC
4K Emissions (Fed./Calif.)
Sales-Weighted 4K
Emissions (Fed./Calif.)
HC
.42/.18
.42/.S6
.67/.40
.84/.33
.64/.37
.S8/.23
.55/.30
.72/.24
.60/.35
.54/.41
.67/.21
.4S/.23
.40/.24
.30/.24
.47/.23
.93/.17
1.30/.20
.61/.65
.S7/.78
.50/.55
1.40/.23
CO
5.62/2.47
9.52/5.64
6.35/4.43
7.07/3.35
13.00/9.50
11.00/1.81
10.35/7.90
6.90/4.15
6.89/3.53
12.74/5.18
9.25/2.59
5.61/3.26
6.23/2.37
7.14/2.54
9.11/3.25
12.18/1.99
6.09/1.47
4.72/4.98
5.52/6.80
6.80/8.90
13.00/3.51
NOx
1.50/1.30
1.70/1.25
2.00/1.41
1.65/1.19
1.51/1.84
1.60/1.28
1.18/1.45
1.46/1.75
1.58/1.58
1.30/1.50
1.57/1.13
1.60/1.15
1.49/1.10
1.56/1.20
1.75/1.30
1.77/1.17
1.70/1.20
1.75/1.36
1.50/2.00
2.00/1.70
1.80/1.14
Total =
Adjusted Totals* =
Federal:
California:
1983 Federal Targets:
HC
.0039/.0017
.0101/.0087
.0095/.0057
.0050/.0020
.0161/.0144
.0149/.0059
.0215/.0117
.0567/.0189
.0805/.0469
.1230/.0934
.0167/.0052
.0045/.0023
.0042/.0025
.0024/.0019
.0012/.0006
.0183/.0033
.0034/.0005
.0029/.0031
.0030/.0041
.0008/.0009
.0249/.0041
.423S/.2378
.61
.34
CO
.0523/.0230
.2294/.1359
.0902/.0629
.0424/.0201
.3276/.2394
.2827/.0465
.4047/.3089
.5430/.3266
.9239/.4734
2.9010/1.180
.2302/.0645
.0555/.0323
.0654/.0249
.0578/.0206
.0237/.0085
.2399/.0392
.0158/.0038
.0222/.0234
.0293/.0360
.0109/.0142
.2314/.0625
6.779/3.1466
9.80
4.55
NOx
.0140/.0121
.0410/.0301
.0284/.0162
.0099/.0071
.0381/.0464
.0411/.0329
.0461/.0567
.1149/.1377
.2119/.2119
.2960/.3416
.0391/.0281
.0158/.0114
.0156/.0116
.0126/.0097
•0046/.0034
.0349/.0230
.0044/.0031
.0082/.0064
.0080/.0106
.0032/.0027
.0320/.0203
1.0198/1.0230
1.47
1.48
.49
5.50
1.40
Adjusted for sales not represented.
-------
Table L-6
1980 California and Federal 4K Emission Levels
Sales-Weighted at 1983 Engine Mix (Fed. /Calif .)
4
6
8
%
1980*
HC CO NOx
cyl. .1143/.0324 1.3867/.4120 .2480/.1764
cyl. .1336/.0575 1.6898/.8390 .33257. 3184
cyl. .36437.2538 6.7232/.3.326 .8936/.9850
1983 Federal Estimated Targets:
% Reduction from 1980 Federal to 1980 California:
Reduction from 1980 Federal necessary for 1983 Federal:
1983*
HC CO NOx
.1135/.0322 1.3775/.4093 .2464/.1752
.2545/.1095 3.2187/1.5981 .6333/.6065
.2565/.1878 4.7346/2.3423 .6293/.6930
.49 5.5 1.4
HC CO NOx
-48% -53% -2.6%
-21% -41% -7.3%
* From Table L-5.
*'*• Using adjusted sales fractions from Table L-4.
-------
with a hypothetical 5 percent engine fuel economy penalty due to
emission control in 1983, an overall increase in fleet fuel economy
will occur due to vehicle improvements. Furthermore, it has been
demonstrated that shifts in consumer buying habits toward more fuel
efficient product lines in California have made the 1980 LOT
California fleet more fuel efficient than the 1980 Federal fleet by
6.5 percent !J/8_/ Shifts in market share do not directly offset the
effects attributable to tighter emission standards, however, and it
is viewing the engine as <* separate entity with which the analysis
will be primarily concerned.
NHTSA stated that it believed a 3 percent average fleet fuel
economy improvement to be a reasonable estimate of the effect' of
electronic engine controls, presuming introduction of a variety of
controls of varying levels of sophistication in 1982.j>/ NHTSA also
quoted a range of 3-5 percent improvement which they believed
attributable to electronic engine controls.^/ The EPA Staff's
analysis of electronic engine controls - in particular their
availability and fuel economy potential - is presented in Reference
7. Based upon a review of the published literature, the Staff must
conclude that electronic engine controls of moderate complexity (in
particular electronic spark advance and EGR modulation) will be
sufficient to offset the worst case 4 percent fuel economy impact
of the 1983 emission standards on the average engine. (Note that
there are HC and CO control strategies which do not degrade full
economy, e.g., increasing oxidation catalyst volume and noble metal
loading. In conjunction with electronic spark advance to handle
transient warm-up/catalyst light-off conditions, the 1983 HC and CO
standards will not be difficult to attain.) The EPA staff, how-
ever, takes issue with NHTSA's claim that the use of EEC's will
become prevalent in 1982. This claim is too optimistic. The only
electronic controls used in 1981 LDT's are a few California engine
families (AMC, Ford), and to presume a 100 percent fleet conversion
to EECs the very next model year (1982) is unreasonable. The EPA
staff foresees a phased introduction of EECs into the Federal fleet
beginning in 1982 and reaching universal incorporation by 1985.
The motivating force for this introduction will be the need to
maintain fuel economy with decreasing emission standards. This is
the basis for EPA's position that no net fuel economy impact is
attributable to the 1983 LDT standards, because moderately complex
EECs will be incorporated at that time to allow simultaneous
compliance with both emission and fuel economy standards. In
short, no degradation in fuel economy is anticipated.
The staff also concludes that the exact control strategies
used for 1983 will depend upon the fuel economy standards mandated
by NHTSA. In the absence of a sizeable NOx reduction, the use of
three-way catalysts and feedback carburetors is not necessary from
an emissions standpoint. Compared to an engine with conventional
EGR, however, a three-way system is more fuel efficient. With the
statutory NOx reductions scheduled for 1985, the staff believes
that three-way catalysts, feedback carburetors, and full electronic
controls will be universally used in 1985 and later light-duty
-------
trucks. These will be basically the same systems which are already
used on the 1981 passenger car fleet. Recognizing the fuel economy
benefits of the three-way system, it is reasonable to conclude that
this technology — available today on a mass production basis -
could, but most probably will not, be applied in 1983. On a
fleetwide basis, electronic controls with infinitely variable
calibration potential for spark timing and EGR modulation should
alleviate any potential fuel economy penalty without the need for
three-way systems, although that option does exist. These elec-
tronic controls need not be as complex as the more elaborate
systems (e.g., General Motors' C-4) and hence need not be as
expensive to manufacture.
In summary, the staff considers that the worst possible
scenario for 1983 light-duty truck fuel economy involves absolutely
no technological innovation. This "quick-fix" approach - with due
regard to the relative stringency of the 1983 standards - would at
most result in an engine fuel economy penalty of 4.0 percent. Even
for this case, a net increase in fleet fuel economy is antici-
pated. However, tried and proven new control technologies are on
the market today which can alleviate any engine fuel economy
penalty attributable to the emission standards. More advanced
technology - anticipated for 1985 at the latest - can increase fuel
economy over and above the 1982 level. The staff believes that the
majority of manufacturers will incorporate controls consistent with
today's technologies (e.g., oxidation catalysts, EGR, air injec-
tion) in addition to electronic controls for spark timing and EGR
modulation to maximize performance and fuel economy. Some manu-
facturers (Volkswagen, for example) may elect to introduce three-
way catalyst/feedback carburetor systems in 1983 to enhance fuel
economy. In conjunction with NHTSA's future fuel economy stan-
dards, the 1983 fleet will be more fuel-efficient than today's. As
discussed above, technological modifications to achieve the 1983
emission standards need not degrade the fleet fuel efficiency of
light-duty truck engines.
4. Recommendation
No net fuel economy penalty is attributable to the proposed
light-duty truck emission standards within the range of cost
effective technological options open to the industry.
-------
References
_!/ "Analysis of California 2% Audit Data," by T. Tesoriero,
March, 1980, available in the Public Docket No. OMSAPC-79-2.
2/ "Preliminary Regulatory Analysis of Light-Duty Truck Fuel
Economy Standards, Model Year 1982-85,", National Highway
Traffic Safety Administration, December, 1979.
_3/ "Optimization of Automotive Engine Calibration for Better Fuel
Economy - Methods and Applications," U.E. Auiler, et al., SAE
Paper 770076.
tjj "Spark Ignition Engine Fuel Economy Control Optimization -
Technique and Procedure," T. Trella, SAE Paper 790179.
5/ "An Approach to a Standard Engine Management System for 1983
~~ and Beyond," by R. Matney, et al., SAE Paper 800470, February,
1980.
6J "Light Truck Average Fuel Economy Standards: Standards for
1982 Model Year," 45 FR 20874, March 31, 1980.
7/ "Electronic Engine Controls - Availability, Durability, and
~ Fuel Economy Effects on 1983 and Later Model Year Light-Duty
Trucks," by T. Cox, Z. Diatchun, T. Nugent, Draft EPA Tech-
nical Report, June 1980.
8/ "Passenger Car and Light Truck Fuel Economy Trends Through
~~ 1980," by J.D. Murrell, et.al., SAE Paper NO. 800853.
-------
M. Issue: Environmental Impact
1. Summary of the Issue
In the draft Regulatory Analysis which accompanied the NPRM,
EPA had, among other things, examined the environmental impact of
the proposed regulations, the cost effectiveness of the proposal,
and possible alternative actions. One reason for preparing these
analyses was to satisfy the requirements of Executive Order 12044,
Improving Government Regulations. A number of comments on the
accuracy and completeness of the Regulatory Analysis materials were
received during the comment period.
2. Summary of the Comments
Commenters identified what they felt were two missing elements
in the analysis of the NPRM. These missing elements were in the
cost effectiveness analysis and the requirement for preparation of
pollutant specific studies.
Although EPA had evaluated the overall benefits and cost
effectiveness of the rulemaking, comments were directed at a lack
of analysis of individual components of the proposal. Such analy-
ses, commenters believed, were required by Section 317 of the Clean
Air Act and by Executive Order 12044. Included in the analysis of
individual components (i.e., emission standards, diesel crankcase
control, revised definition of useful life, revised durability
provisions, revised maintenance provisions, SEA, PCA) should be a
consideration of alternatives to those components.
Cummins engine company commented on EPA's failure to follow
what it considered to be the standard setting framework of the 1977
Clean Air Act amendments for those light-duty trucks in the 6,000-
8,500 Ib. GVW range. Cummins interpretation of this standard
setting framework is presented in Figure M-l. In that framework,
the Administrator has two paths to follow in establishing target
standards. The first is via the percent reduction requirements of
Section 202(a)(3)(A)(ii). This section establishes the requirement
for a 90 percent reduction in HC and CO for 1983, and a 75 percent
reduction in NOx for 1985. The second path is via the "pollutant
specific studies" of Section 202(a)(3) (E). On the basis of these
studies, the Act allows the Administrator to change the above
percent reduction standards for Section 202(a).
Once the target standards were established, the Administrator
has the further option of a temporary revision to those standards
based upon considerations of technology, cost and fuel penalty. If
made, such a revision would be effective for a 3-year period,
beginning 4 years after adoption.
Cummins indicated that in its interpretation, this entire
process was to have been completed by December 31, 1978, for
standards effective in 1983. Every 3 years after that the process
1*41
-------
FIGURE M-1
CLEAN AIR ACT
202(a)(3)(A), (B), and (E)
As Interpreted By Cummins
->—Health Effects Study
or
Target Standards
based on 90%, 75%
reductions from
uncontrolled levels
\
Temporary Revised
Standards
based on available
technology, energy
and cost penalties
Target Standards
based on health
•effect requirements
effective 4 years after
adoption
every 3 years
-------
would be repeated, with an opportunity to change or vevise the
standards once again. Cummins felt that, rather than simply
relying on the statutory percentage reduction requirements, EPA was
to conduct its standard setting while cognizant of the basic
underlying purpose of the Clean Air Act to "promote the public
health and welfare."
EPA received a number of specific comments (principally from
Ford) on the methodology used to conduct its environmental analy-
sis. Ford cited what it found to be incorrect treatment of Cali-
fornia regions (which, of course, are subject to a distinct set of
emission standards). Ford stated that "(t)he calculated air
quality gains associated with this incorrect assumption are greater
than the benefits associated with the proposed regulations." Ford
also felt that the EPA analysis supported the position that most of
the air quality gains expected in future years would occur whether
or not the NPRM were finalized. Ford stated that redefined useful
life and in-use vehicle durability testing were not proven to have
a significant air quality impact. Ford indicated its belief that
the benefits were so small as to be unnoticeable.
Ford went on to prepare air quality projections of its own,
using the EPA MOBILE 1 and rollback models. The details of that
analysis can be found in the Ford submission of October 11, 1979,
beginning at page 4 of Section VI. The results supported Ford's
position that the light-duty truck NPRM, as a whole, would have
little air quality benefit compared to the air quality improvements
already projected due to other source reductions, and that redefi-
nition of useful life also has little or no associated benefit.
Ford's analysis also indicated that a very moderate I/M program (10
percent stringency and no mechanics training) would provide a
greater air quality benefit than those achieved by the proposed LDT
standards in 1983.
In its air quality analysis, Ford also examined the proposed
idle standards. Because of the regional nature of oxidant air
pollution, Ford felt there was no basis for an HC idle standard.
Relative to CO, Ford attempted to assess the effect of that pro-
posed standard by applying speed correction factors to the EPA
emission factors. These showed that the rollback model would
predict a greater improvement in air quality for areas characte-
rized by lower average traffic speeds than for areas characterized
by higher average traffic speeds. Ford drew the implication that
this "suggests that the idle mode for CO may not be as important as
EPA seems to have implied ..."
Ford indicated that when the exhaust HC emission rates are
broken down into methane and non-methane fractions, methane is
shown to be a significant portion of the total. This methane
fraction, Ford believes, should be excluded from exhaust measure-
ments because it is photochemically inert.
Comment from GM and others pointed out that tamper-resistant
-------
components would be appearing on light-duty trucks. EPA'«= para-
meter adjustment regulations take effect in 1981 and 1982, and the
impact of these regulations on in-use vehicle emissions needs to be
accounted for in the analysis of benefits.
The Alaska Department of Environmental Conservation and the
EPA Alaska Air Coordinator drew attention to the fact that emission
rates for CO are known to be highly temperature dependent. They
therefore felt that the projected emissions and air quality bene-
fits for CO would not be fully realized in Alaskan regions.
3. Analysis of the Comments
The EPA staff shares the desire of commenters for a more
detailed cost effectiveness analysis of the components of the
rulemaking. However, as was noted in the preamble to the NPRM
(44FR 40793, July 12, 1979), there was a lack of data on which to
base such analysis. During the comment period, EPA has endeavored
to collect sufficient information to examine costs and benefits
associated with the several elements of the rulemaking. Although
many of the difficulties facing the technical staff at the time of
the NPRM are still present, an analysis of the elements of the
final rulemaking has been prepared and can be found in Chapter VII
of the Regulatory Analysis.
The Cummins interpretation of the 1977 Clean Air Act amend-
ments regulatory scheme is not a new one to the EPA technical
staff. This issue was also raised during the rulemaking action for
1984 and later heavy-duty engines.^/ Concerning the "pollutant
specific studies" required under Section 202 (a) (3) (E), it is the
technical staff's opinion that, although not specifically identi-
fied by that name, these studies have in fact been completed -
first as part of the draft Regulatory Analysis, and now in the
final Regulatory Analysis. The Regulatory Analysis takes a compre-
hensive look at the environmental impact (i.e. health and welfare
effects via the ambient air quality standards) of the control
strategy represented by the statutory 90 percent reductions. It
also considers alternate standards of varying stringency, along
with such aspects as costs and cost effectiveness.
The EPA technical staff does not subscribe to the dual path
options as described by Cummins. In reality, if that approach
were followed, the Congressionally prescribed percentage reduction
targets would have no meaning and might never be used. It is most
unlikely that the pollutant specific study would identify precisely
the same percentage reductions contained in the Amendments.
Rather, the process is a sequential one, wherein the pollutant
specific study is used to evaluate the level or control desired
by Congress from an environmental health viewpoint to determine its
appropriateness. The study would consider the effects of the
statutory percentage reductions, and if these were desireable and
adequate in the context of emission reduction and air quality
benefits, then those percentage reductions would be retained.
-------
The Administrator could also decide to change the standard. Making
this judgment would include consideration of such things as
the gain or loss of benefits associated with more stringent or less
stringent standards.
We do not view the pollutant specific studies as intended by
Congress to double check the standard which it chose, or as a
condition precedent to promulgating standards at the statutory
levels. If that were the case, we would expect Congress to have
first applied the concept to light-duty vehicles, not heavy-duty
trucks. It is there that the greatest environmental and economic
impacts of a given standard occur. Rather, the statutory percen-
tage reductions serve as a starting point which represents the
desire of Congress. The pollutant specific studies provide a means
of changing those standards in the case of a significant environ-
mental need to do so. This is consistent with the Clean Air Act
provisions for the use of the pollutant specific study, which is,
in the wording of Section 202 (a) (3) (E) (ii), for "changing any
standard prescribed...." according to the statutory reductions.
The technical staff acknowledges that Congress had originally
envisioned the above process as taking place in 1978, for standards
applicable in the 1983 model year. For further discussion of this
matter see "Issue E - Leadtime."
We will now turn to the comments on EPA's environmental
assessment methodology. The technical staff agrees with the
comments concerning analysis of California regions. Since Cali-
fornia maintains its own emission standards, the staff recommends
analyzing only non-California regions in the final Regulatory
Analysis.
The technical staff also agrees with those commenters who
pointed out that most of the overall air quality improvements
projected for future years would occur whether or not new light-
duty truck regulations are adopted. However, the staff strongly
disagrees with the conclusion of commenters that the light-duty
truck regulations are therefore unnecessary, or of minor value.
For indeed, the overall air quality improvements can be broken down
into any number of smaller increments related to various source
categories - such as light-duty vehicles, heavy-duty trucks, and
various stationary source categories. The fact that each of these
categories produces incremental benefits which look small compared
to the sum total cannot be used to dismiss the import of those
categories, or the total benefit would soon be gone. The total
array of air quality improvements resulting from other known
control strategies here forms a backdrop or environment in whichi
the light-duty truck regulations play a role commensurate with
their relative emission rates.
Ford claimed that its air quality analysis indicated that
redefined useful life and in-use durability testing had no proven
benefit. That analysis was based upon Ford's interpretation and
extension of work in the draft Regulatory Analysis. Ford used the
-------
"worst case" assumptions about in-use deterioration in its assess-
ment. These worst case assumptions minimize the effects of
extended useful life and over-state the relative importance of I/M
programs. In preparing the final Regulatory Analysis, the tech-
nical staff has found that the worst case assumptions from the draft
analysis were unnecessarily pessimistic, and has developed a more
realistic approach. The final Regulatory Analysis also will
analyze the benefits of individual components of the package and
evaluate their cost effectiveness.
Concerning the need for idle standards, the technical staff
has recommended elsewhere (Issue D - Idle Test and Standards) that
the HC idle standard be deleted from the final rule. The staff has
further recommended retention of the CO idle standard. The ration-
ale for that decision is set forth in the above issue analy-
sis. The argument here advanced by Ford using speed correction
factors to the mobile source emission factors is tenuous at best.
Ford itself hesitates to draw anything more than a "suggestion"
from it. The speed correction factors are not valid means of
estimating idle emissions. They apply to grams per mile emission
rates, which clearly do not fit the idle case where the vehicle is
emitting while standing still.
In agreement with the suggestion by Ford, the air quality
analysis will be done on the basis of non-methane fractions of
hydrocarbon emissions. This change makes the mobile source emis-
sion factors consistent with the non-methane hydrocarbon emission
rates used for stationary source categories in the emission inven-
tors data base. Cost effectiveness calculations, on the other
hand, have historically been based on total hydrocarbons. Since
total hydrocarbons are being reduced by the regulations, and since
a prime function of cost effectiveness is to provide a measure for
comparison with other strategies, total hydrocarbons will continue
to be used for cost effectiveness.
The GM comments concerning the need to consider the pre-1983
impact of parameter adjustment regulations for light-duty trucks
are well taken. The final Regulatory Analysis will include correc-
tions to the emission factors to incorporate the effect of param-
eter adjustment.
Comments concerning CO problems in Alaskan regions were also
valid. The emission factors used for the air quality assessment
are not appropriate for areas of persistent low temperatures.
Therefore, Alaskan regions, along with California and high-altitude
regions, will be excluded from the air quality assessment. This
does not imply a lack of benefit in Alaskan or high-altitude areas
from this rulemaking. If, for example, manufacturers had to reduce
cold-start emissions in order to meet their target emission levels,
then Alaska might experience a substantially greater benefit than
predicted. The staff does not believe that this rulemaking need
include provisions or analysis for non-FTP test conditions since
this problem is being looked at elsewhere. Further discussion
-------
can be found in Issue N: Special Exemptions.
4. Recommendations
The staff recommends that the following changes be included in
the final Regulatory Analysis.
1. Expand the cost effectiveness analysis to include anal-
ysis of individual elements of the rulemaking.
2. Exclude California, Alaska, and high-altitude regions
from the air quality assessment.
3. Consider non-methane hydrocarbon emissions in pre-
paring the air quality impact analysis.
4. Incorporate the effect of light-duty truck parameter
adjustment regulations into the 1981 and 1982 model year emission
factors.
-------
References
For discussion see "Summary and Analysis of Comments to the
NPRM: 1983 and Later Model Year Heavy-Duty Engines, Proposed
Gaseous Emission Regulations," December 1979, p. 268.
-------
N. Issue: Special Exemptions
1. Summary of the Issue
EPA has received comments in two areas that do not fall within
the boundaries of major issue categories covered in this document.
Both deal with special situations which the commenters believe
require special treatment. These issues will be analyzed as
Special Exemptions.
2. Summary of the Comments
a. Exemption From All Emission Regulations of Trucks up to
9000 Pounds Licensed as Agriculture Vehicles
The New Mexico Cattle Grower's Association (NMCGA) has requested an
exemption from the new rulemaking and a modification of existing
regulations for trucks up to 9,000 pounds that are licensed as
agricultural vehicles. The following issues represent the Associ-
ation's major justifications for exemption:
(1) Insignificant environmental degradation due to vast
emission dispersion.
(2) Increased vehicle cost.
(3) Increased vehicle fuel consumption.
(4) Decreased vehicle reliability.
(5) Decreased vehicle power output.
(6) Catalytic Converters.
(a) Dual bulk storage requirement to handle unleaded
fuel.
(b) Increased demand on unleaded fuel supplies.
(c) Conversion of other farm vehicles to unleaded fuel.
(d) Range fire potential.
(7) Electronic ignition problems of unreliability and unre-
pairability.
(8) Additional costs associated with these regulations will
be borne by the consumer.
t
b. Low Temperature CO Emission Problem and Applicability of
the Federal Test Procedure (FTP) to Areas With Typically
Lower Operating Temperatures
The Alaska Department of Environmental Conservation (ADEC) and
the Municipality of Anchorage have indicated concern over the
proposed control measures and applicability of the FTP for areas
-------
that typically experience low-temperature winters. ADEC specif-
ically addresses the need for certification or deterioration
rate testing to be conducted over a full range of temperatures.
Since Alaska's nonattainment problems are primarily related
to cold starts, the Department indicates that if low cold-start
test temperatures are not instituted, the added vehicular cost for
emission control will not result in the stated 26 percent reduction
of CO in Alaska or in other states experiencing wintertime CO
problems.
Input from the EPA Air and Hazardous Waste Division and
its Alaska Air Coordinator is in agreement indicating that the
proposed regulations do not control cold temperature CO emissions
or reduce CO emission by the aforementioned figure and should
therefore be revised to control CO emissions at temperatures below
75° F.
3. Analysis of Comments
a. EPA's analysis of the NMCGA comments has indicated
that direct application to non-agricultural vehicles can be made
for a majority of the issues. EPA therefore, does not consider
them adequate justification for regulatory exemption of the vehi-
cles in question (light-duty trucks). Two concerns, however, the
potential of catalyst ignited range fires and the requirement for
dual bulk fuel storage capability, are directly related and
deserve analysis.
Studies conducted principally by the U.S. Forest Service have
indicated that catalyst equipped vehicles exhibit no significantly
higher fire potential than pre-catalyst vehicles. This is because
catalyst skin temperatures are similar to pre-catalyst exhaust
system temperatures. These simple statements are not meant to
evade the issue. EPA realizes the potential for some increased
hazard in range or range like situations but we believe that
control of the problem is the responsibility of the manufacturer.
The failure or destruction rate of this equipment is an additional
problem to be solved via the design route not through regulation
revision.
Concerning the need to acquire bulk storage capability for
storing unleaded fuel, it is important to note that catalysts are
already in use on virtually all light-duty trucks as a result of
existing regulations. Finalization of this rulemaking, therefore,
will have no significant impact on that situation. Since the
number of agricultural vehicles.in this category is increasing each
year it would appear reasonable to assume that a dual bulk storage
capability is being acquired on an ever increasing number of
ranches. Furthermore, the high potential that ranchers' private
automobiles utilize unleaded fuel would serve as added incentive to
acquire that capacity. It is also a logical assumption that the
ever increasing demand for unleaded fuel will result in a propor-
tional supply increase.
-------
Other concerns, if these vehicles were exempt, include
possible vehicle resale for full-time on road use. Emission
control at that point would be nonexistent and it is unreason-
able to assume, due to time and dollar costs involved, that control
devices would be installed prior to resale. In addition, a produc-
tion run to produce exempt vehicles would result in increased
"special production" costs that may negate potential savings
arrived at by catalyst or other emission control device elimina-
tion.
b. EPA recognizes"the low temperature emission problem and
that of applicability of the FTP to Alaska and further acknowledges
existence of these problems in other areas of the country as well.
The Agency is currently in the process of developing a cold room
for the simulation of low temperature test conditions (20°F).
Testing, to be conducted, will provide data required for the
development of FTP guidelines for low temperatures.
The strategy for low temperature emission control assumes the
use of electronic engine control systems. These systems provide
the potential of greater benefit in the colder areas with Alaska
being the case of point.
Since EPA recognizes that the current FTP does not accurately
measure low temperature emissions it is understood that the esti-
mated reductions of CO may not apply to Alaska or other low temper-
ature areas. With the development of low temperature test proce-
dures the Agency is, however, moving in a positive direction in its
drive to reduce low temperature CO emissions and will continue to
review potential alternative solutions to the issue presented.
4. Recommendations
No changes in the regulations are recommended in response to
these issues.
------- |